Investigation of the Effects of Heat on

Bone Tissues to Inform Forensic

Analysis

Meghan Raechel Mckinnon

Faculty of Health and Medical Sciences, the

University of Adelaide

A thesis submitted for the degree of Doctor of

Philosophy
June, 2021
 Table of Contents
Abstract............................................................................................................................................. 1
Declaration ........................................................................................................................................ 3
Acknowledgements ........................................................................................................................... 4
Chapter 1: Introduction ...................................................................................................................... 5
1.1 DNA inheritance and analysis .................................................................................................. 7
1.2 Bone microstructure and taphonomy ...................................................................................... 12
1.3 Ethical considerations ............................................................................................................ 15
1.4 Primer design and application ................................................................................................ 17
1.5 Thesis scope .......................................................................................................................... 18
1.6 References ............................................................................................................................. 20
Chapter 2: A Review of the Current Understanding of Burned Bone as a Source of DNA for Human
Identification ................................................................................................................................... 29
Statement of Authorship .............................................................................................................. 30
2.1 Introduction ........................................................................................................................... 31
2.2 Sample collection................................................................................................................... 32
2.3 Sample preparation ................................................................................................................ 33
2.4 DNA Extraction ..................................................................................................................... 34
2.5 Areas for expansion ............................................................................................................... 34
2.6 References ............................................................................................................................. 36
Chapter 3: DNA Extraction from Incinerated Bone .......................................................................... 38
3a. Validation of porcine primers ................................................................................................. 39
3a.i Introduction ...................................................................................................................... 39
3a.ii Methods ........................................................................................................................... 40
3a.iii Concluding statement ......................................................................... 44
3b: Comparison of Bone Demineralisation Procedures for DNA Recovery from Burned Remains 45
Statement of Authorship .......................................................................................................... 46
3b.1 Introduction ..................................................................................................................... 47
3b.2 Methods ........................................................................................................................... 48
3b.3 Results & discussion ........................................................................................................ 49
3b.4 Conclusion ....................................................................................................................... 51
3b.5 References ....................................................................................................................... 51
Chapter 4: A Comparison of Crystal Structure in Fresh, Burned and Archaic Bone – Implications for
Forensic Sampling ........................................................................................................................... 53
Statement of Authorship .............................................................................................................. 54
 4.1 Introduction ........................................................................................................................... 55
4.2 Methods................................................................................................................................. 56
4.3 Results................................................................................................................................... 57
4.4 Discussion ............................................................................................................................. 60
4.5 References ............................................................................................................................. 62
Chapter 5: Effects of Thermal Insult on Bone Tissue as Observed by Micro Computed Tomography 64
Statement of Authorship .............................................................................................................. 65
5.1 Introduction ........................................................................................................................... 66
5.2 Methods................................................................................................................................. 67
5.3 Results................................................................................................................................... 68
5.4 Discussion ............................................................................................................................. 70
5.5 Conclusion............................................................................................................................. 72
5.6 References ............................................................................................................................. 73
Chapter 6: General Discussion and Conclusions .............................................................................. 74
6.1 Discussion ............................................................................................................................. 75
6.2 References ............................................................................................................................. 81
Appendix ........................................................................................................................................ 83
Appendix i. Histological analysis of incinerated bone .................................................................. 84
References ............................................................................................................................... 89
Appendix ii. Awards and other achievements ............................................................................... 90
 Abstract
Following an incineration event bone is often the only surviving tissue, making the skeleton vital for

obtaining a positive identification of the deceased. Discriminating features such as age, sex, height,

and antemortem injuries can often be observed in skeletal remains and used as secondary identifiers in

forensic examination. Understanding the effects of incineration on bone tissues and how this may

impact skeletal markers is essential to improving standard methods of anthropological profiling for

forensic identification. In addition to these secondary identifiers, bone provides a protective barrier for

DNA which is highly individualising and hence is considered a primary scientific method of

identification. Due to the structural degradation of bone, and the loss of organic material, obtaining a

viable DNA sample from incinerated bone is problematic. There is a need for improved DNA

extraction protocols for these incinerated samples, as well as methods of triaging multiple samples to

facilitate an optimal outcome.

The primary aim of this research was to study heat-induced changes in bone microstructure to inform

forensic evidence recovery and analysis. This was achieved by employing a range of techniques

including x-ray powder diffraction (XRPD), micro computed tomography (microCT), and scanning

electron microscopy (SEM) with energy dispersive x-ray (EDX). As previous research has suggested

heat-induced changes in bone mineral resemble long term diagenetic changes, it was hypothesised

that ancient DNA extraction methods may be appropriate for modern incinerated samples. This theory

was tested by comparing demineralisation techniques designed for ancient DNA extraction to simpler

techniques used for fresh samples. The quantity of DNA recovered was compared between methods to

assess the optimal method for extraction from burned samples. As bone crystallinity has an impact on

both DNA preservation and the ability to extract DNA from mineralised tissue, XRPD was used to

assess heat induced changes in hydroxyapatite crystal size and structure across a range of temperature

intervals. SEM and EDX were then used to visually assess these changes and provide basic elemental

analysis. In the final part of this project, MicroCT was used to create 3D reconstructions of

1
incinerated bone samples to visually and quantitatively assess changes to bone porosity as a result of

thermal insult.

The application of ancient bone demineralisation techniques was not shown to increase DNA yield

from incinerated bones, with a shorter demineralisation period and retention of EDTA supernatant

producing optimal results. XRPD analysis provided a possible explanation for this, showing that heat-

induced changes to bone hydroxyapatite were far more impactful than the diagenetic changes seen in

ancient bone, which could not be distinguished from unburned modern samples. Similarly marked

heat-induced changes in bone structure were observed using MicroCT, where bone porosity was

greatly decreased at high temperatures. This research highlights the variable nature of heat induced

changes in bone which could prove valuable in prioritising sampling based on the likelihood of

obtaining identifying information such as DNA.

2
 Declaration

I certify that this work contains no material which has been accepted for the award of any other

degree or diploma in my name, in any university or other tertiary institution and, to the best of my

knowledge and belief, contains no material previously published or written by another person, except

where due reference has been made in the text. In addition, I certify that no part of this work will, in

the future, be used in a submission in my name, for any other degree or diploma in any university or

other tertiary institution without the prior approval of the University of Adelaide and where

applicable, any partner institution responsible for the joint award of this degree.

I acknowledge that copyright of published works contained within this thesis resides with the

copyright holder(s) of those works.

I also give permission for the digital version of my thesis to be made available on the web, via the

University’s digital research repository, the Library Search and also through web search engines,

unless permission has been granted by the University to restrict access for a period of time.

I acknowledge the support I have received for my research through the provision of an Australian

Government Research Training Program Scholarship

Meghan Mckinnon

Signed: Date: 23.06.21

3
 Acknowledgements

Firstly, I’d like to thank my co-supervisors Ellie Simpson and Maciej Henneberg, who have been with

me since I was a lowly undergrad student – I really appreciate you sticking with me for all these

years. Special thanks to my primary supervisor Denice Higgins; you always know when I am in need

of encouragement or motivation, I could not have completed this thesis without you! Secondly, I

would like to thank Dr Qiuhong Hu and Mr Xuanzhao Pan from the Faculty of Engineering for not

even flinching when I asked to put burned cow bone in their very expensive equipment, Dr Jennifer

Young for showing unparalleled patience whilst teaching advanced laboratory techniques to a near

novice, the wonderful staff at Adelaide Microscopy, especially Ruth Williams and Agatha Labrinidis

who made MicroCT fun. Completing a PhD can be a stressful and often lonely experience – luckily

for me I had some wonderful officemates who made it so much better. For this I want to thank my

fellow PhD students; Angela Gurr, work mum extraordinaire who has kept me well fed for the past

three years, and Florence Lees who always has the solution to every problem. Also Dr Jaliya

Kumaratilake for all his kind-hearted support, and the Forensic Odontology group for accepting me as

one of their own - teeth really aren’t so different from bone! Finally, big shout out to my family and

friends including Ashleigh and Mauro who were kind enough to lend me their workshop and sit

around for hours incinerating bones (even if they didn’t really know why), my amazing grandparents,

brother, and of course my mum and dad who allowed me to live at home (rent free) for far longer than

any parents should have to. Thank you all for your love and support, I couldn’t have done it without

you!

4
Chapter 1: Introduction

5
The formal identification of a deceased person is a legal requirement in most jurisdictions. It also

provides closure and comfort to the family of the deceased individual. Positive identification is not

always an easy task – aside from logistical constraints such as decomposition, displacement and

commingling of remains, death is a sensitive area for many individuals. Cultural and religious

considerations must also be taken into account. Many belief systems do not approve autopsies, for

example Jewish law Halacha requires remains to be buried immediately with some rare exceptions.

Hindu practices require cremation of remains within 24 hours of death, and Islamic law requires

remains be buried within this same timeframe to avoid loss of any parts of the body, which is

considered sacrilegious. Repatriation of historical Aboriginal remains is another sensitive task faced

by forensic anthropologists, who attempt to provide as much information as possible in order to

identify the home country of the deceased individual so they may be repatriated. We as forensic

professionals have a legal, ethical and humanitarian duty to ensure all human remains are granted

identity and justice after death.

Forensic human identification, be it of victims of foul play, unexpected natural death, manmade

disaster or simple misadventure relies at the most basic level on individual uniqueness. Without the

inherent variation that exists in the anatomy, physiology and cultural aspects of every human, we

would not be able to return identity to those who have lost it or had it taken from them. However, it

is essential that we understand not only how individuals differ, but also how they do not. The

skeleton is an anatomical feature that we all share; when all other traditional identifying features

have been lost, the bones and teeth remain. Dental records, fingerprints and DNA are considered

primary identifiers, meaning they are scientifically validated techniques (Schuliar & Knudsen,

2012). In the absence of primary identifiers, secondary traits such as tattoos, scars, or an

anthropological profile may be used to support identification. In some cases, anthropological

assessment of pathologies, abnormalities and injuries can provide highly individualising information

(De Boer et al., 2020). However, if none of these features are present, analysis of the skeleton alone

may not be sufficient for a positive identification. The mineralised nature of bone, however, protects

and preserves one of the most discriminating aspects of human variation: DNA (Pinhasi et al., 2015).

6
1.1 DNA inheritance and analysis

DNA is a double helical compound consisting of an alternating sugar and phosphate backbone with

attached nucleotide bases. Much like a binary coding system, the relatively simple alteration of base

pairs (Guanine – Cytosine, Adenine – Thymine) provides an immensely complex blueprint instructing

the development of an entirely unique organism (Lee et al., 1994). In the case of humans, these

blueprints can be found within either the nuclei of cells or within cellular mitochondria. Nuclear DNA

(nDNA) is tightly packaged within 46 chromosomes, 44 of which are autosomal non-sex

chromosomes, whilst two are sex chromosomes. Autosomal chromosomes exist in pairs, with

offspring inheriting one homologous chromosome maternally and the other paternally (Butler, 2009).

A specific physical location on a chromosome is known as a “locus”, plural “loci”, and variations of a

specific locus are called “alleles”. Determining which of the parents’ homologs will be passed on to

the offspring is a random process; this is primarily how a unique genetic profile is created. Meiotic

crossing over is another aspect of genetic inheritance that increases genotypic variation. When

chromosome homologs align during meiosis, DNA can be cut, shuffled between pairs and re-annealed

to create a new, unique homologous pair. Although crossing over does not always occur in gamete

production, the random allocation of parental chromosomes does. The combination of these processes

results (with the exception of monozygotic twins) in a highly individualised genetic profile (Latham

& Miller, 2018, Baranzini et al., 2010, Kaye, 2010). More recent studies have shown that examination

of methylation patterns in DNA will even allow differentiation between identical twins (Stewart et al.,

2015, Vidaki et al., 2013).

Although individual profiles are essential to establishing identity, there are some cases where forensic

professionals are interested in establishing kinship between genetic samples. In these cases,

mitochondrial DNA (mtDNA) can be used in place of, or in addition to nDNA (Fig. 1.1). As the name

suggests, mtDNA is housed within the mitochondria of eukaryotic cells. It is theorised that the

development of mitochondria and their unique genome is a result of endosymbiosis, whereby

ancestral eukaryotic cells internalised energy producing bacterial prokaryotes (Zimorski et al., 2014).

This resulted in formation of double-membraned organelle like mitochondria and chloroplasts, with

the inner

7
membrane derived from the ancestral bacteria and the outer from the host-cell. As the mitochondria of

all energy producing cells using aerobic processes were once isolated prokaryotic cells, they still

possess self-contained, isolated genomes in the form of a single circular chromosome (Kodama &

Fujishima, 2010). This means the mitochondrial genome now exists as multiple copies in each

eukaryotic cell, unlike the single nDNA genome housed within the nucleus. MtDNA has a unique

germline inheritance, with all genetic material required for organelle function (primarily production of

adenosine triphosphate in the citric acid cycle) passed on predominantly by the mother (Thornton et

al., 2014). This differs significantly from the recombinant nature of nDNA, which is inherited

biparentally (Budowle et al., 2003). Although the mtDNA genome is not subject to the genetic

randomisation of parental genes, it is far more susceptible to evolutionary alterations and mutations

than nDNA. Base substitutions in mammalian mtDNA genomes have been calculated to occur at a

rate of 0.02 substitutions per base pair per million years, which exceeds that of the nDNA genome

tenfold (Brown et al., 1979). This makes the mitochondrial genome ideal for evolutionary genetic

studies.

Fig 1.1. Comparison of structural and compositional differences between mitochondrial and nuclear

DNA. Figure created with BioRender.com.

8
Despite the rapid rate of genomic evolution, this occurrence is slow enough that inter-generational

genomes are unaffected. This means mtDNA is still very useful for establishing genealogy and

kinship. In disaster victim identification (DVI), mtDNA can be used to identify maternally related

family members following cases where related individuals are known to be amongst the deceased

(vehicular crashes, house fires etc.) (Holland et al., 1993). However, the lack of individuality in the

mitochondrial genome is a major caveat in many cases as the discriminatory power of mtDNA is

much lower than nDNA, and profiles obtained can be matched to a larger pool of individuals (Bär et

al., 2000, Giles et al, 1980). For this reason, in closed DVI situations where a majority of victims are

documented (e.g., flight manifest) kinship is usually established using nDNA profiles voluntarily

provided by families, as was used to aid in identifying many of the victims of Malaysian Airlines

flight MH17 (Smith & Mann, 2015). However, in cases where remains are subjected to extreme

degradation, mtDNA can have some advantage over nDNA due to the higher number of

mitochondrial genomes and the circular molecular structure, which is more stable than nuclear helical

DNA (Montelius & Lindblom, 2012, Graw et al., 2000). These factors result in higher survivability of

mtDNA under challenging conditions, making it particularly useful for establishing kinship of

degraded, skeletal remains (Coble et al. 2004). In Chapter 3 of this thesis, where alternate methods of

DNA extraction from bone are assessed, I focused on nDNA, with the understanding that mtDNA has

much greater survivability than nDNA, hence if nDNA can be retrieved mtDNA must also be

available in equal or greater measures.

In order to compare and contrast two genetic profiles we must first identify regions of inter-individual

variability, and DNA polymorphisms are ideal for this. These variations can be in the form of single

bases, multiple bases, and repeated sequences. The insertions and deletions of individual bases are

known as “sequence polymorphisms” whilst those that involve the insertion of multiple bases are

known as “length polymorphisms”. Population genetic studies have mapped a large number of

locations at which polymorphisms occur, which forensic professionals can then use to isolate and

compare sequences from unknown and known samples. Initially, restriction fragment length

polymorphisms (RFLP) served as the primary method of isolating genetic variation. This involves the

9
use of probes and simple restriction endonucleases to tag and digest the sample, after which the

remaining fragments can be visualised using electrophoresis to provide a unique profile based on the

varying length of fragments. Although still useful for genomic studies and general variation analysis,

in forensic science it has largely been replaced by the use of polymerase chain reaction (PCR).

The advent of PCR had a massive impact on the field of genetics, particularly in forensic science.

Unlike RFLP, PCR can be used to isolate and amplify miniscule quantities of DNA from extremely

degraded samples in a fraction of the time (1-2 days vs 6-8 weeks) (Butler, 2009). PCR works by

identifying a DNA sequence containing the chosen polymorphism, isolating it with the use of primers

and exponentially replicating the target sequence multiple times creating enough of the target product

for various forms of analyses (e.g., gel electrophoresis). Real-time quantitative PCR (qPCR) takes this

process a step further and allows the number of target fragment copies to be measured in real time.

This is achieved by use of fluorescent labelling (e.g SYBR® Green dye or TaqMan® probes) which

binds to double-stranded DNA or primers. As the number of fragment copies increases with each

amplification cycle, so too does the fluorescence. The point at which the fluorescence reaches a

detectable level above background noise is known as the CT (Cycle Threshold) or Cq (Cycle

quantification) value (Karlen et al., 2007). In order to quantify the DNA concentration of an unknown

sample, the Cq values of known samples are plotted against the log of DNA concentration to create a

dosage-dependent standard curve. Unknown sample Cq values are then plotted against this curve to

extrapolate the concentration of the sample. This is an example of absolute qPCR quantification,

which is most commonly used in forensic analysis as opposed to relative quantification (used to

measure gene expression).

Early methods of DNA profiling only allowed analysis of long repeat sequences, classified as

“minisatellites”. The first minisatellites to be used were variable number tandem repeat (VNTR)

markers, which were often analysed using RFLP in a process known as DNA

“fingerprinting” (Butler, 2009). With increased genome mapping and the development of highly

specific amplification techniques such as PCR, smaller markers soon replaced VNTRs. The term

“microsatellite” is commonly used to describe this family of smaller markers, which have a much

higher power of discrimination than the earlier minisatellites (Lee et al., 1994).

10
In addition to the highly polymorphic nature of microsatellites, these markers are also less affected by

DNA strand breakage and are more readily retrieved from degraded samples due to their shorter

length.

Today the most commonly used minisatellites are short tandem repeat (STR) markers, with STR-

based human identification kits used in most forensic laboratories (Schichman et al., 2002).

Traditional STR kits targeted amplicons of 100 to 450 base pairs in length, however for highly

degraded DNA even fragments of this length may not be available. To overcome this “mini STRs”

were developed which involved moving the primers targeting the amplicon as close as possible to the

repeat sequence. While this improves the likelihood of recovering a useful DNA profile it can still be

problematic for highly degraded samples. A possible alternative in these situations are single

nucleotide polymorphisms (SNP), which as their name suggests, are a variation at a single base. SNPs

which are largely bi-allelic are of interest to the forensic community for a multitude of reasons; the

amplicons targeted for SNPs can be very short and within a population a SNP can be assigned a

frequency of occurrence. Additionally, phenotypic predictions can be made based on varied allelic

combinations. One of the shortcomings of SNPs is finding a multiplex assay that can amplify enough

SNP targets for robust, discriminatory results (Butler, 2009). This, and the fact that extensive STR

databases currently exist, means SNPs are unlikely to replace STRs in a forensic criminal

investigation context in the foreseeable future.

The use of DNA for identification of human remains in a forensic setting most often occurs when a

body is not visually identifiable. That is, the body is damaged or decomposed to an extent that it

would not be reliable to confirm identity based on the opinion of a friend or relative viewing the facial

appearance of the deceased. When taphonomic degradation is extensive, for example following

advanced diagenesis, prolonged burial or incineration, mineralised tissue such as teeth and bones may

be all that remains for genetic analysis. Teeth are particularly well suited for genetic identification as

the DNA-rich pulp is protected by dense enamel (Balasse, 2002, Bryant et al., 1996). However, teeth

are also easily displaced due to various factors such as disease, aging and diagenesis (Higgins &

Austin, 2013), and in such cases bone may be the only substitute. DNA retrieval from these

11
mineralised tissues is often more complicated than from fresh material. Archaeologists and

palaeobiologists have extensively researched DNA recovery from diagenetic skeletal remains, and

some of this research can be applied to modern forensic casework (Latham and Miller, 2019).

Substantially fewer studies have investigated the changes that occur in incinerated bone, and how this

may affect DNA recovery. There is a particular need for this research in Australia where bushfires are

frequent and severe, with climate change likely to aggravate the situation in future years (Lucas et al.,

2007).

1.2 Bone microstructure and taphonomy

Successful forensic analysis of skeletal tissues that have been degraded by heat requires an

understanding of what specific changes take place within the bone during the burning process. For

DNA related investigations this knowledge is especially important to inform which bone to target and

how best to liberate DNA from within these tissues. At the most basic level, bone can be divided into

two main structural forms that can be observed macroscopically; trabecular and cortical. The

trabecular (or cancellous) structure of bone tissue is produced by organic osteoids laid around spaces

filled with red bone marrow. The osteoid is permeated by hydroxyapatite so that ultimately bone

trabeculae are built of collagen fibres strengthened by hydroxyapatite (Trueman et al., 2008). By

comparison cortical bone consists of Haversian systems that surround small blood vessels, and of

osteocyte spaces (Cummaudo et al., 2019, Vaughan et al., 2012, Qiu et al., 2003). Some research has

shown that at lower temperatures (<250 °C) the circularly organised lamellar layers of Haversian

systems undergo significant changes, primarily the broadening of Haversian canals as observed with

SEM (Chadefaux & Reiche, 2009). This reorganisation of bone internal structure is likely a primary

driver of changing porosity. Bone is also often described as primary or secondary, which simply refers

to the time and manner in which it is formed – primary bone mostly occurs during early development

and repair, whereas secondary bone (also known as Haversian bone) is either deposited onto existing

bone surfaces or onto sites where pre-existing bone has been resorbed (Hillier & Bell, 2007).

Both trabecular and cortical bone can be further divided into organisationally different

microstructures that are particularly useful for species differentiation. Woven bone is a primary tissue

12
commonly found in fetal bones and repair sites. In early development, as well as fracture repair,

woven bone is laid down in a rapid and disorderly fashion to provide a temporary scaffolding

structure for mature bone to develop on (Mulhern & Ubelaker, 2012). Comparatively lamellar bone is

a mature tissue that consists of consecutive layers of thin lamella deposited alongside parallel layers

of collagen that can form either primary or secondary bone. Fast growing mammals including cows,

pigs and sheep can also present a combination of woven and lamellar tissue known as fibrolamellar

bone which is largely absent in humans and other primates (Martiniaková et al., 2007). This tissue

consists of alternating layers of woven and lamellar structures which, although somewhat

mechanically inferior to primary lamellar bone, can be laid down far more rapidly (Mulhern &

Ubelaker, 2012). A subset of this tissue where lamellae are organised in a brick-like pattern is known

as plexiform fibrolamellar bone, which is highly characteristic of large mammals (Burr & Martin,

1989). The presence of plexiform bone in animal models is likely an important structural difference

that must be considered when making comparisons to humans.

Bone diagenesis and/or weathering results in chemical and mechanical destruction of bone. A

consequence of this can be a severe decrease in bone density, which has been linked to the physical

protection of endogenous DNA (Pinhasi et al., 2015, Collins et al., 2002). Additionally, decreased

bone integrity dramatically increases the likelihood of microbial invasion, and contamination of target

DNA. It has been suggested that ancient bones exposed to prolonged diagenetic processes may share

similar structural composition to modern incinerated bones, however research supporting this is

insubstantial (Piga et al., 2009, Stiner et al., 1995). Understanding these similarities (or lack thereof)

is of particular importance to how DNA extraction from incinerated bone may be approached,

especially with regards to the demineralisation step, which has been shown to be vital for ancient

samples. As the name implies, DNA is slightly acidic due to the phosphates in the molecular

backbone. This means nucleic acids project a slight negative charge which allows them to bind to

ionised substances such as bone hydroxyapatite (HAp) (Busse et al., 2009, Paget et al., 1992). In

addition to removal of calcium (which is a PCR inhibitor), demineralisation is essential for liberating

small quantities of DNA bound to HAp crystallites. In archaic samples most cellular DNA has been

lost, and a prolonged period of demineralisation is vital to release as much endogenous DNA as

possible. The comparison of diagenetic ancient bone and modern incinerated bone is a common

theme throughout this thesis.

13
Changes in bone integrity, and in turn the likelihood of obtaining viable DNA, can be assessed to

some degree by microscopic and macroscopic investigation. Whilst it is often visually obvious if a

sample is too degraded for analysis (i.e., calcined bone), caution must be taken as the depth of internal

damage is not always visible from the exterior. This is where microscopic changes such as presence

or absence of nucleated cells, osteon integrity and micro fracturing can be used to give some

indication of DNA survival (Edson et al., 2013, Singh et al. 2013, Alers et al., 1999, Cattaneo et al.,

1999). Collagen content has also been linked to DNA yield, with some studies suggesting an

increased affinity for DNA binding as the collagen becomes mineralised (Campos et al., 2011,

Gӧtherstrӧm et al., 2002, Collins et al., 1995). Histological analysis using both demineralised/stained

(i.e., haemotoxylin and eosin, masson’s trichrome) and unstained (i.e., resin embedded slices) tissue

can show these changes (Higgins et al., 2015, 2013), however decalcifying, cutting and mounting

highly degraded or incinerated samples can be challenging.

As histological analysis requires a certain amount of sample destruction, it is not ideal for cases

where target material is delicate or limited. In such cases, non-invasive alternatives such as Micro

Computed Tomography (MicroCT) should be considered. MicroCT is a less traditional method for

examination of structural integrity, however with some refinement it could prove a valuable screening

tool for examination of forensic samples. This could be achieved by employing volumetric analysis

(bone density, bone porosity etc.) of incinerated samples using three dimensional MicroCT

reconstructions (Kuhn et al., 1990, Sandholzer et al., 2013). Utilising MicroCT in this way could

provide both visual and quantitative insight into heat-induced changes in bone without the need for

extensive sample preparation or destruction.

Research throughout this thesis attempts to clarify if changes to the microstructure of incinerated

modern bones are comparable to those of ancient bone using a holistic approach to explore both large-

scale and molecular taphonomy. Chapter 3 directly investigates the benefits of ancient DNA

extraction demineralisation techniques on modern incinerated samples, whilst Chapter 4 investigates

how HAp crystallites change during burning (compared to ancient Bison bone). Chapter 5 uses a

combination of MicroCT and SEM to compare histological and microstructural changes in remains

incinerated at various temperatures.

14
1.3 Ethical considerations

When conducting research on biological samples, a range of factors must be taken into consideration,

especially with model selection. In some regards working on post-mortem tissues is less ethically

constrained than research on live subjects however, as samples are often exposed to destructive

processes, animal substitutes are generally preferable to human. As the outcome of this research is

aimed at informing processes applicable to human tissues any substitute must represent humans as

closely as possible whilst still taking various practical factors into consideration. In many cases, the

need for similarity in bone macro/micro structure is overcome by factors such as developmental rate

of an induced disease, or ease of handling. This is why small rodents such as domestic mice and rats

are commonly used in osteoarthritis research – surgically induced osteoarthritis can be closely

monitored in these models, and lesions develop much more rapidly than in larger species (Bendele et

al., 2001). In the past, research on orthopaedic implants has used larger animals to varying degrees of

success. The main structural difference between human bone tissue and that of large animals like

sheep, cows and pigs is the presence of primary plexiform bone – this type of bone tissue is

characteristic of these fast-growing animals and is mostly absent in human bones (Hillier & Bell,

2007). In immature animals, plexiform tissue can be located throughout entire long bones. For this

reason, it is imperative that samples be collected only from mature individuals to increase the

likelihood of obtaining secondary Haversian bone that is more comparable between species.

Domestic sheep (Ovis aries), have increased in popularity for orthopaedic implant studies as they

have macroscopically similar bone structure to humans, are easily available for research and are easier

to handle than some other large animals (for example cows and pigs) (Pearce et al., 2007). Although

macroscopically similar, the microstructure of cortical sheep bone consists of irregularly shaped

haversian canals as opposed to the dense, consistently shaped canals of human bone (De Kleer 2006,

Mosekilde et al., 1987). Cows are also occasionally used in research; however their large size make

them generally unfavourable for in vivo studies. Like sheep, they possess cortical bone consisting of

irregularly shaped Haversian canals (Pearce et al., 2007). Aerssens et al., (1998) compared the

structure and composition of various laboratory animals including sheep, cows, pigs and rats. The

15
findings suggest that although pigs (along with dogs) shared more biomechanical similarities with

humans, cows were the intermediary, surpassing rats and sheep in terms of bone comparability.

Domestic pigs (Sus scrofa) are arguably the best substitute for human models due to the similarity of

their body size and bone composition both macro and microscopically. Like humans, pigs possess

densely organised, regularly shaped Haversian canals (although porcine trabecular bone is slightly

denser) (Thorwarth et al., 2005). Furthermore, the genome of pigs is relatively similar to humans,

which makes them similarly viable substitutes for DNA based research (Fadista et al. 2008, Hart et

al., 2007). Although pigs are not ideal for in vivo studies due to their high growth rates, large size and

noisy, aggressive natures, this issue is negated when using scavenged post-mortem tissues, making

them ideal for forensic research.

No single species can perfectly replace human samples, and ultimately the animal models I used for

my research were selected based on a combination of ethical and logistical considerations. Human

samples would have been ideal, but due to the destructive nature of the research this was not possible.

A porcine model was the obvious substitute for the DNA-based study covered in Chapter 3 due to the

genomic similarities between humans and pigs. Additionally, pig carcases are readily available and

easy to acquire, and as they are deceased none of the issues involved with housing and handling living

animals apply here. Model selection was also based on availability, as was the case for the study

presented in Chapter 4 where ancient bone was compared to modern samples. As this study was

comparing and contrasting the effect of different taphonomic influences (diagenesis vs incineration) it

was vital that samples were comparable. Ancient bone samples are much harder to obtain than

fresh/modern material, and as I was only able to acquire ancient bovine samples (Bison antiquus),

domestic cow (Bos taurus) was the only modern counterpart readily available. Despite some minor

difference in density and organisation, mammalian bone microstructure is inherently similar.

However, as different regions of the skeleton are afforded different levels of protection by the

surrounding tissue thickness (Buikstra and Swegle, 1989), and the distribution of tissue differs

between cows, pigs and humans, heat-induced changes to bone microstructure may not be directly

comparable to humans. This is an unavoidable caveat of using animal models for forensic research.

16
1.4 Primer design and application

Opting to use a porcine model was not without challenges, primarily the lack of established primer

sets. Primers are single-stranded DNA fragments designed to be complimentary to nucleotide regions

on target DNA (Butler et al., 2009). During PCR, DNA strands are separated, allowing primers to

bind and amplify the target material. Two primers, one forward and one reverse, are required to target

both complementary DNA strands (Garibyan & Avashia, 2013). As no suitable published porcine

nuclear primers were available, they had to be specifically developed in house and validated before I

could apply them to my research. When designing primers various factors such as the number and

composition of base pairs (GC to AT ratio), individual specificity of target fragment, annealing

temperature and secondary structure of the molecule must be considered to optimise PCR efficacy.

The optimal length of a standard primer is generally between 18 and 24 base pairs; any longer than

this and hybridization times become excessive, whilst much shorter primers can result in

amplification of non-target material (Wu et al., 1991). When extracting and amplifying DNA from

highly degraded samples, short primers are preferable as they are more sensitive to fragmented DNA

(Swango et al., 2006).

As GC bases contain more hydrogen bonds than adenine and thymine, increasing GC content can

increase primer stability (Kämpke et al., 2001). It is generally suggested that a primer should consist

of between 45% to 60% GC, and (if possible) end with either a G or C base “clamp” to promote

binding at the 3’ end of the strand (Dieffenbach et al., 1993, Sheffield et al., 1989). However,

including too many G or C repeats near the 3’ end should be avoided as this can lead to self-

hybridisation, and subsequent primer-dimer (Abd-Elsalam, 2003). Hairpins are another example of a

secondary primer structure that should be avoided. These structures form when two separate regions

on the same primer are complementary, which allows the strand to fold back and anneal to itself

forming a loop or “hairpin” (Fan et al., 2019). Online databases like primer-BLAST (Ye et al., 2012)

can be used to ensure primer specificity, assess its potential for self-binding and identify temperatures

at which problematic secondary structures may form (Thornton & Basu, 2011). Online databases and

primer design tools can also be used to establish the melting point of a primer (Tm). Ideally, annealing

17
should occur at temperatures around 5 °C below Tm, and making sure this does not occur at

particularly low temperatures is essential to prevent non-specific binding (Rychlik et al., 1990).

Ideally there should be no more than 2 °C difference in Tm between primers. Although Tm can be

calculated based on the GC content (higher temperatures needed for higher GC content), realistically

this temperature may vary in the reaction itself (Bustin & Huggett, 2017). To achieve the best result,

the primer reaction (e.g., thermocycling conditions) may require further optimisation.

Along with DNA quantity, DNA quality can also be assessed using qPCR by the amplification of two

DNA target strands of different lengths (Swango et al., 2006, Niederstätteret al., 2007). qPCR primers

are designed to select target areas of DNA that do not vary between individuals, only occur once in

the genome, and are species specific. If the long target fragment can be successfully amplified, this

indicates the DNA strands have not been severely degraded or broken (Nygard et al., 2007, Alonso et

al., 2005). Common techniques of isolating desired DNA sequences involve the use of fluorescent

dyes or probes. Fluorescent dyes such as SYBR® green bind to double-stranded DNA molecules,

allowing the amount of DNA present after each thermocycle to be recorded (Higgins et al., 2015).

Comparatively, fluorescent probes such as those used in TaqMan® assays bind to primer-specific

sequences; this results in cumulative florescence of probes after each amplification cycle which is

proportional to the quantity of DNA synthesised (Arya et al., 2005). Although TaqMan® assays are

more product-specific, they are also more complex and costly (Tajadini et al., 2014). For this reason,

I opted to use SYBR® green for my research.

1.5 Thesis scope

In the wake of destructive post-mortem processes, DNA and skeletal material are intrinsically linked,

and the degradation of one often results in loss of the other. With this in mind, I have taken a

multidisciplinary approach to investigate the following aspects of the bone-DNA relationship. The

bulk of this thesis is presented in a series of published manuscripts with the exception of the

introduction (Chapter 1), and discussion and conclusions (Chapter 6). The first manuscript (Chapter

2) is a literature review compiling established information on DNA extraction and amplification from

incinerated samples. In addition to providing a general overview of some of the pitfalls and

shortcomings of the currently available methods, I outline areas where knowledge is lacking, and the

potential for further research.

18
The second manuscript (Chapter 3) explores direct DNA yield from burned bones using various

methods of decalcification. Although using human samples for this study would have been ideal,

ethical and logistical constraints made it impossible. Porcine primers were developed as an

alternative, however this presented a new set of challenges. To address this, chapter 3 is separated

into two parts; section 3a, which includes non-published but still relevant information relating to how

porcine primers were developed and tested, and section 3b, which includes the published study

exploring decalcification techniques using these primers. Initially it was believed that degraded

modern bone samples should be exposed to prolonged decalcification prior to extraction in the same

way that archaic bones are. With the previous study showing that bone calcite crystal structure only

changes dramatically after 600 °C, I hypothesised that using extraction techniques designed for

archaic bones may actually be too harsh for many burned samples, and a decreased EDTA

incubation period could improve DNA yield. Additionally, research suggested that retaining digested

supernatant could prevent the loss of vital DNA (with the increased potential for contamination)

(Fondevila et al., 2008, Loreille et al., 2007), which I also decided to test. These altered extraction

protocols were tested on a range of burned and unburned bone samples, following which absolute

quantification using qPCR was used to assess the success of each method.

The third manuscript (Chapter 4) expands on the link between bone hydroxyapatite and DNA by

investigating how heat-induced degradation can affect this link. A combination of x-ray powder

diffraction (XRPD), scanning electron microscopy (SEM) and energy dispersive x-ray (EDX) were

used to visualise and quantify changes to bone crystallite structure at various temperature intervals.

Here I also discuss the potential use of the aforementioned techniques as screening tools to

differentiate various bone types (e.g., ancient/fresh, porcine/bovine). The primary aim of this study

was to assess if hydroxyapatite (HAp) crystallites would increase or decrease in size as a result of

heating. As DNA extraction methods are designed to remove excess calcium from highly crystallised

ancient samples, comparing the crystallite composition of ancient and modern incinerated bones

could indicate if these methods are also appropriate for modern samples.

The fourth and final manuscript (Chapter 5) investigates the use of MicroCT as a method for

assessing incinerated bone compared to standard SEM investigation. MicroCT provided information

19
on changes to bone porosity that could not be fully realised using SEM alone, additionally these

changes could be both quantified and visually observed using 3D reconstruction. Ultimately the

aimof this study was to assess what information MicroCT could provide on changes occurring in

burned bone, and also to compare it to changes observed using SEM analysis. Histological analysis

was also explored, however decalcification and subsequent cutting of incinerated samples was

largely unsuccessful. The limited findings of this analysis have been included in appendix i.

1.6 References

Abd-Elsalam, K 2003, ‘Bioinformatic tools and guideline for PCR primer design’, African Journal of

Biotechnology, vol. 2, no. 5, pp. 91-95.

Aerssens, J, Boonen, S, Lowet, G & Dequeker, J 1998, 'Interspecies differences in bone composition,

density, and quality: Potential implications for in vivo bone research', Endocrinology, vol. 139, no. 2,

pp. 663-670.

Alers, J, Krijtenburg, P, Vissers, K, & Van Dekken, H 1999, ‘Effect of bone decalcification

procedures on DNA In situ hybridization and comparative genomic hybridization: EDTA is highly

preferable to a routinely used acid decalcifier.’, Journal of Histochemistry & Cytochemistry, vol. 47,

no. 5, pp. 703-709.

Alonso, A, Martin, P, Albarran, C, Garcia, P, Fernandez de Simon, L, Jesus Iturralde, M, et al. 2005,

'Challenges of DNA profiling in mass disaster investigations', Croatian Medical Journal, vol. 46, no.

4, pp. 540-548.

Arya, M, Shergill, I, Williamson, M, Gommersall, L, Arya N & Patel, H 2005, ‘Basic principles of

real-time quantitative PCR’, Expert Review of Molecular Diagnostics, vol. 5, no. 2, pp. 209-219.

Balasse, M 2002, ‘Reconstructing dietary and environmental history from enamel isotopic analysis:

time resolution of intra‐tooth sequential sampling’, International Journal of Osteoarchaeology, vol.

12, no. 3, pp. 155-165.

Bär, W, Brinkmann, B, Budowle, B, Carracedo, A, Gill, P, Holland, M, et al. 2000, ‘DNA

commission of the international society for forensic genetics: guidelines for mitochondrial DNA

typing’, Forensic Science International, vol. 110, no. 2, pp. 79-85.

20
Baranzini, S, Mudge, J, Van Velkinburgh, J, Khankhanian, P, Khrebtukova, I & Miller, N 2010,

‘Genome, epigenome and RNA sequences of monozygotic twins discordant for multiple sclerosis’,

Nature, vol. 464, no. 7293, pp. 1351.

Bendele, A 2001, ‘Animal models of osteoarthritis’, Journal of Musculoskeletal Neuronal

Interactions, vol. 1, no. 4, pp. 363-76.

Brown, W, George, M & Wilson, A 1979, ‘Rapid evolution of animal mitochondrial DNA’,

Proceedings of the National Academy of Sciences, vol. 76, no. 4, pp. 1967-1971.

Bryant, J, Koch, P, Froelich, P, Showers, W & Genna, B 1996, ‘Oxygen isotope partitioning between

phosphate and carbonate in mammalian apatite’, Geochimica et Cosmochimica Acta, vol. 60, no. 24,

pp. 5145-5148.

Budowle, B, Allard, M, Wilson, M & Chakraborty, R 2003, ‘Forensics and mitochondrial DNA:

applications, debates, and foundations’, Annual Review of Genomics and Human Genetics, vol. 4, no.

1, pp. 119-141.

Buikstra, J & Swegle M 1989, ‘Bone modification due to burning: experimental evidence’, In:

Bonnichsen, R & Sorg, M (eds), Bone Modification, pp. 247-258, University of Maine, Maine.

Burr, D & Martin, R 1989, ‘Errors in bone remodeling: toward a unified theory of metabolic bone

disease’, American Journal of Anatomy, vol. 186, no. 2, pp. 186-216.

Busse, B, Hahn, M, Soltau, M, Zustin, J, Püschel, K, Duda, G, et al. 2009, ‘Increased calcium

content and inhomogeneity of mineralization render bone toughness in osteoporosis: mineralization,

morphology and biomechanics of human single trabeculae’, Bone, vol. 45, no. 6, pp. 1034-1043.

Bustin, S & Huggett, J 2017, ‘qPCR primer design revisited’, Biomolecular Detection and

Quantification, vol. 14, pp. 19-28.

Butler, J 2009, Fundamentals of forensic DNA typing, Academic press.

Campos, P, Craig, O, Turner-Walker, G, Peacock, E, Willerslev, E & Gilbert, M 2011, ‘DNA in

ancient bone–Where is it located and how should we extract it?’, Annals of Anatomy-Anatomischer

Anzeiger, vol. 194, pp. 7-16.

21
Cattaneo, C, Dimartino, S, Scali, S, Craig, O, Grandi, M & Sokol, R 1999, ‘Determining the human

origin of fragments of burnt bone: a comparative study of histological, immunological and DNA

techniques’, Forensic Science International, vol. 102, no. 2-3, p. 181-191.

Chadefaux, C & Reiche, I 2009, ‘Archaeological bone from macro-to nanoscale: heat-induced

modifications at low temperatures’, Journal of Nano Research, vol. 8, pp. 157-172.

Coble, M, Just, R, O’Callaghan, J, Letmanyi, I, Peterson, C, Irwin, J, et al. 2004, ‘Single nucleotide

polymorphisms over the entire mtDNA genome that increase the power of forensic testing in

Caucasians’, International journal of legal medicine, vol. 118, no. 3, pp. 137-146.

Collins, M, Riley, M, Child, A & Turner-Walker, G 1995, ‘A basic mathematical simulation of the

chemical degradation of ancient collagen’, Journal of Archaeological Science, vol. 22, no. 2, pp.

175-183.

Collins, M, Nielsen–Marsh, C, Hiller, J, Smith, C, Roberts, J, Prigodich, R, et al. 2002, ‘The survival

of organic matter in bone: a review’, Archaeometry, vol. 44, no. 3, pp. 383-394.

Cummaudo, M, Cappella, A, Giacomini, F, Raffone, C, Màrquez-Grant, N & Cattaneo, C 2019,

‘Histomorphometric analysis of osteocyte lacunae in human and pig: exploring its potential for

species discrimination’, International Journal of Legal Medicine, vol. 133, no. 3, pp. 711-718.

De Boer, H, Obertová, Z, Cunha, E, Adalian, P, Baccino, E et al. 2020, ‘Strengthening the role of

forensic anthropology in personal identification: position statement by the board of the Forensic

Anthropology Society of Europe (FASE).’, Forensic Science International, vol. 315, pp. 110456.

De Kleer, V 2006, ‘Development of bone’, In: Sumner-Smith, G (ed), Bone in Clinical

Orthopaedics, pp. 1-80, WB Saunders Co, Philadelphia.

Dieffenbach, C, Lowe, T & Dveksler, G 1993, ‘General concepts for PCR primer design’. PCR

Methods and Applications, vol. 3 no. 3, pp. S30-S37.

Edson, J, Brooks, E, McLaren, C, Robertson, J, McNevin, D, Cooper, A, et al. 2013, ‘A quantitative

assessment of a reliable screening technique for the STR analysis of telogen hair roots’, Forensic

Science International: Genetics, vol. 7, no. 1, pp. 180-188.

22
Fadista, J, Nygaard, M, Holm, E, Thomsen, B, Bendixen, C 2008, ‘A snapshot of CNVs in the pig

genome’, PLOS One, 3, pp. e3916.

Fan, H, Wang, J, Komiyama, M & Liang, X 2019, ‘Effects of secondary structures of DNA templates

on the quantification of qPCR’, Journal of Biomolecular Structure and Dynamics, vol. 37, no. 11, pp.

2867-2874.

Fondevila, M, Phillips, C, Naveran, N, et al., 2008, ‘Case report: identification of skeletal remains

using short-amplicon marker analysis of severely degraded DNA extracted from a decomposed and

charred femur’, Forensic science international: Genetics, vol. 2, no. 3, pp. 212-218.

Garibyan, L & Avashia, N 2013, ‘Research techniques made simple: polymerase chain reaction

(PCR)’, The Journal of Investigative Dermatology, vol. 133, no. 3, pp. e6.

Giles, R, Blanc, H, Cann, H & Wallace, D 1980, ‘Maternal inheritance of human mitochondrial

DNA’, Proceedings of the National academy of Sciences, vol. 77, no. 11, pp. 6715-6719.

Götherström, A, Collins, M, Angerbjörn, A & Lidén, K 2002, ‘Bone preservation and DNA

amplification’, Archaeometry, vol. 44, no. 3, pp. 395-404.

Graw, M, Weisser, H, & Lutz, S 2000, ‘DNA typing of human remains found in damp environments’

Forensic Science International, vol. 113, no. 1, pp. 91-95.

Hart, E, Caccamo, M, Harrow, J, Humphray, S, Gilbert, J, Trevanion, S, et al. 2007, ‘Lessons learned

from the initial sequencing of the pig genome: comparative analysis of an 8 Mb region of pig

chromosome 17’, Genome biology, vol. 8, no. 8, pp.1-12.

Higgins, D & Austin, J 2013, ‘Teeth as a source of DNA for forensic identification of human remains:

a review’, Science & Justice, vol. 53, no. 4, pp. 433-441.

Higgins, D, Rohrlach, A, Kaidonis, J, Townsend, G & Austin, J 2015, 'Differential nuclear and

mitochondrial dna preservation in post-mortem teeth with implications for forensic and ancient DNA

studies', Plos One, vol. 10, no. 5, pp. 1-17.

23
Higgins, D, Kaidonis, J, Townsend, G, Hughes, T & Austin, J 2013, ‘Targeted ampling of cementum

for recovery of nuclear DNA from human teeth and the impact of common decontamination

measures’, Investigative Genetics, vol. 4, no. 1, pp.18.

Hillier, M, & Bell, L 2007, ‘Differentiating human bone from animal bone: a review of histological

methods’, Journal of Forensic Sciences, vol. 52, no. 2, pp. 249-263.

Holland, M, Fisher, D, Mitchell, L, Rodriquez, W, Canik, J, Merril, C, et al. 1993, ‘Mitochondrial

DNA sequence analysis of human skeletal remains: identification of remains from the vietnam war’,

Journal of Forensic Sciences, vol. 38, no. 3, pp. 542-553.

Kämpke, T, Kieninger, M & Mecklenburg, M 2001, ‘Efficient primer design algorithms’,

Bioinformatics, vol. 17, no. 3, pp. 214-225.

Karlen, Y, McNair, A, Perseguers, S, Mazza, C & Mermod, N 2007, ‘Statistical significance of

quantitative PCR’, BMC Bioinformatics, vol. 8, no. 1, pp. 131.

Kaye, D 2010, ‘Probability, individualization, and uniqueness in forensic science evidence’,

Brooklyn Law Review, vol. 75, no. 4, pp. 8.

Kodama, Y & Fujishima, M, 2010, in International Review of Cell and Molecular Biology.

Kuhn, J, Goldstein, S, Feldkamp, L, Goulet, R, Jesion, G 1990, ‘Evaluation of microcomputed

tomography system to study trabecular bone structure’, Journal of Orthopaedic Research, vol. 8, no.

6, pp. 833-842.

Latham, K & Miller, J 2019, ‘DNA recovery and analysis from skeletal material in modern forensic

contexts’, Forensic Sciences Research, vol. 4, no. 1, pp. 51-59.

Lee, H, Ladd, C, Bourke, M, Pagliaro, E & Tirnady, F 1994, ‘DNA typing in forensic science. I.

Theory and background’, The American Journal of Forensic Medicine and Pathology, vol. 15, no. 4,

pp. 269-282.

Loreille, O, Diegoli, T, Irwin, J, et al., 2007, ‘High efficiency DNA extraction from bone by total

demineralization’, Forensic Science International: Genetics, vol.1, pp. 191-195.

24
Lucas, C, Hennessy, K, Mills, G & Bathols, J 2007, Bushfire weather in southeast Australia: recent

trends and projected climate change impacts, Consultancy Report prepared for The Climate Institute

of Australia, CSIRO Marine and Atmospheric Research, Melbourne.

Martiniaková, M, Grosskopf, B, Omelka, R, Dammers, K, Vondráková, M & Bauerová, M 2007,

‘Histological study of compact bone tissue in some mammals: a method for species determination’,

International Journal of Osteoarchaeology, vol. 17, no. 1, pp. 82-90.

Montelius, K & Lindblom, B 2012, "DNA analysis in disaster victim identification.", Forensic

Science, Medicine, and Pathology, vol. 8, no. 2, pp. 140-147.

Mosekilde, L, Kragstrup, J & Richards, A 1987, ‘Compressive strength, ash weight, and volume of

vertebral trabecular bone in experimental fluorosis in pigs’, Calcified Tissue International, vol. 40,

no. 6, pp. 318-322.

Mulhern, D & Ubelaker, D 2012, ‘Differentiating human from nonhuman bone microstructure. Bone

histology’, In An Anthropological Perspective, pp. 109-134.

Niederstätter, H, Köchl, S, Grubwieser, P, Pavlic, M, Steinlechner, M & Parson, W 2007, ‘A modular

real-time PCR concept for determining the quantity and quality of human nuclear and mitochondrial

DNA’, Forensic Science International: Genetics, vol. 1, no. 1, pp. 29-34.

Nygard, A, Jørgensen, C, Cirera, S & Fredholm, M 2007, ‘Selection of reference genes for gene

expression studies in pig tissues using SYBR green qPCR’, BMC Molecular Biology, vol. 8, no. 1,

pp. 67.

Paget, E, Monrozier, L & Simonet, P 1992, ‘Adsorption of DNA on clay minerals: protection against

DNaseI and influence on gene transfer’, FEMS Microbiology Letters, vol. 97, no 1-2, pp. 31-39.

Pearce, A, Richards, R, Milz, S, Schneider, E & Pearce, S 2007, 'Animal models for implant

biomaterial research in bone: A review', European Cells & Materials, vol. 13, pp. 1-10.

Piga, G, Thompson, T, Malgosa, A & Enzo, S 2009, ‘The potential of X‐ray diffraction in the

analysis of burned remains from forensic contexts’, Journal of Forensic Sciences, vol. 54, no. 3, pp.

534-539.

25
Pinhasi, R, Fernandes, D, Sirak, K, Novak, M, Connell, S, Alpaslan-Roodenberg, S, et al. 2015,

‘Optimal ancient DNA yields from the inner ear part of the human petrous bone’, PLOS One, vol. 10,

no. 6, pp. e0129102.

Qiu, S, Fyhrie, D, Palnitkar, S & Rao, D 2003, ‘Histomorphometric assessment of Haversian canal

and osteocyte lacunae in different‐sized osteons in human rib’, The Anatomical Record Part A:

Discoveries in Molecular, Cellular, and Evolutionary Biology, vol. 272, no. 2, pp. 520-525.

Rychlik, W, Spencer, W & Rhoads, R 1990, ‘Optimization of the annealing temperature for DNA

amplification in vitro’, Nucleic Acids Research, vol. 18, no. 21, pp. 6409-6412.

Sandholzer, M, Walmsley, A, Lumley, P & Landini, G 2013, ‘Radiologic evaluation of heat-induced

shrinkage and shape preservation of human teeth using micro-CT’, Journal of Forensic Radiology

and Imaging, vol. 1, no. 3, pp. 107-111.

Schichman, S, Suess, P, Vertino, A & Gray, P 2002, ‘Comparison of short tandem repeat and variable

number tandem repeat genetic markers for quantitative determination of allogeneic bone marrow

transplant engraftment’, Bone Marrow Transplantation, vol. 29, no. 3, pp. 243-248.

Schuliar, Y & Knudsen, P 2012, ‘Role of forensic pathologists in mass disasters.’, Forensic Science,

Medicine, and Pathology, vol. 8, no. 2, pp. 164-173.

Sheffield, V, Cox, D, Lerman, L & Myers, R 1989, ‘Attachment of a 40-base-pair G+ C-rich

sequence (GC-clamp) to genomic DNA fragments by the polymerase chain reaction results in

improved detection of single-base changes’, Proceedings of the National Academy of Sciences, vol.

86, no. 1, pp. 232-236.

Singh, V, Salunga, R, Huang, V, Tran, Y, Erlander, M, Plumlee, P & Peterson, M 2013, ‘Analysis of

the effect of various decalcification agents on the quantity and quality of nucleic acid (DNA and

RNA) recovered from bone biopsies’, Annals of Diagnostic Pathology, Vol.17, no. 4, pp.322-326.

Smith M & Mann M 2015, ‘Recent developments in DNA evidence’, Trends & Issues in Crime and

Criminal Justice, vol. 506, pp. 1.

26
Stiner, M, Kuhn, S, Weiner, S & Bar-Yosef, O 1995, ‘Differential burning, recrystallization, and

fragmentation of archaeological bone’, Journal of Archaeological Science, vol. 22, no. 2, pp.223-237.

Stewart, L, Evans, N, Bexon, K, van der Meer, D & Williams, G 2015, ‘Differentiating between

monozygotic twins through DNA methylation-specific high-resolution melt curve analysis’,

Analytical Biochemistry, vol. 476, pp. 36-39.

Swango, K, Timken, M, Chong, M & Buoncristiani, M 2006, ‘A quantitative PCR assay for the

assessment of DNA degradation in forensic samples’, Forensic Science International, vol. 158, pp.

14-26.

Tajadini, M, Panjehpour, M & Javanmard, S 2014, ‘Comparison of SYBR Green and TaqMan

methods in quantitative real-time polymerase chain reaction analysis of four adenosine receptor

subtypes’, Advanced Biomedical Research, vol. 3.

Thornton, B, Cohen, B, Copeland, W & Maria, B 2014, ‘Mitochondrial disease: clinical aspects,

molecular mechanisms, translational science, and clinical frontiers’, Journal of Child Neurology, vol.

29, no. 9, pp. 1179-1207.

Thornton, B & Basu, C 2011, ‘Real‐time PCR (qPCR) primer design using free online software’,

Biochemistry and Molecular Biology Education, vol. 39, no. 2, pp. 145-154.

Thorwarth, M, Schultze-Mosgau, S, Kessler, P, Wiltfang, J & Schlegel, K 2005, ‘Bone regeneration

in osseous defects using a resorbable nanoparticular hydroxyapatite’, Journal of Oral and

Maxillofacial Surgery, vol. 63, no. 11, pp. 1626-1633.

Trueman, C, Privat, K & Field, J 2008, ‘Why do crystallinity values fail to predict the extent of

diagenetic alteration of bone mineral?’, Palaeogeography, Palaeoclimatology, Palaeoecology, vol.

266, no. 3-4, pp. 160-167.

Vaughan, T, McCarthy, C & McNamara, L 2012, ‘A three-scale finite element investigation into the

effects of tissue mineralisation and lamellar organisation in human cortical and trabecular bone’,

Journal of the Mechanical Behavior of Biomedical Materials, vol. 12, pp. 50-62.

27
Vidaki, A, Daniel, B & Court, D 2013, ‘Forensic DNA methylation profiling—potential

opportunities and challenges’, Forensic Science International: Genetics, vol. 7, no. 5, pp. 499-507.

Wu, D, Ugozzoli, L, Pal, B, Qian, J & Wallace, R 1991, ‘The effect of temperature and

oligonucleotide primer length on the specificity and efficiency of amplification by the polymerase

chain reaction’, DNA and Cell Biology, vol. 10, no. 3, pp. 233-238.

Ye, J, Coulouris, G, Zaretskaya, I, Cutcutache, I, Rozen S, Madden, T 2012, ‘Primer-BLAST: A

tool to design target-specific primers for polymerase chain reaction’, BMC Bioinformatics, vol. 13,

no. 1, pp. 1-11.

Zimorski, V, Ku, C, Martin, W & Gould, S 2014, ‘Endosymbiotic theory for organelle origins’,

Current Opinion in Microbiology, vol. 22, pp. 38-48.

28
 Chapter 2: A Review of the Current

Understanding of Burned Bone as a

Source of DNA for Human

Identification

Published: Mckinnon, M, Henneberg, M, Higgins, D 2021, ‘A review of the current understanding of

burned bone as a source of DNA for human identification’, Science & Justice, doi:

https://doi.org/10.1016/j.scijus.2021.03.006

29
 Statement of Authorship
Title of Paper: A Review of the current understanding of burned bone as a Source of DNA for Human
Identification
Publication Status: ☒ Published

☐ Accepted for Publication

☐ Submitted for Publication

☐ Unpublished & Unsubmitted work written in Manuscript Style

Publication Details: Published in Science & Justice 2021.

Principal Author (Candidate): Meghan Mckinnon

Contribution to the paper: Conceptualization, formal analysis, investigation, writing – original draft,
writing – editing
Overall percentage (%): 85%
Certification: This paper reports on original research I conducted during the period of my Higher
Degree by Research candidature and is not subject to any obligations or contractual
agreements with a third party that would constrain its inclusion in this thesis. I am
the primary author of this paper.

Signature & Date: __ 23.06.21

__________________________________________

Co-Author Contributions
By signing the Statement of Authorship, each author certifies that:
i. the candidate’s stated contribution to the publication is accurate (as detailed above);
ii. permission is granted for the candidate to include the publication in the thesis; and
iii. the sum of all co-author contributions is equal to 100% less the candidate’s stated contribution.
Co-Author: Maciej Henneberg
Contribution to the paper: Writing – review & editing, supervision, project administration, funding
acquisition.

Signature & Date: __ 23.06.21

________________________________

Co-Author: Denice Higgins (secondary author)

Contribution to the paper: Conceptualization, investigation, writing – review & editing, supervision,
project administration, funding acquisition.

23.06.21
Signature & Date: ___________________ ___________________________

30
 Science & Justice 61 (2021) 332–338

Contents lists available at ScienceDirect

Science & Justice

journal homepage: www.elsevier.com/locate/scijus

Review

A review of the current understanding of burned bone as a source of DNA

for human identification
Meghan Mckinnon a, *, Maciej Henneberg a, Denice Higgins b
a
Discipline of Anatomy and Pathology, Adelaide Medical School, the University of Adelaide, Adelaide, Australia
b
Forensic Odontology Unit, Adelaide Dental School, the University of Adelaide, Adelaide, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Identification of incinerated human remains may rely on genetic analysis of burned bone which can prove far
Burned bone more challenging than fresh tissues. Severe thermal insult results in the destruction or denaturation of DNA in
Ancient bone soft tissues, however genetic material may be preserved in the skeletal tissues. Considerations for DNA retrieval
Bone demineralisation
from these samples include low levels of exogenous DNA, the dense, mineralised nature of bone, and the presence
Forensic DNA recovery
of contamination, and qPCR inhibitors. This review collates current knowledge in three areas relating to opti­
DNA extraction protocol
mising DNA recovery from burned bone: 1) impact of burning on bone and subsequent effects on sample
collection, 2) difficulties of preparing burned samples for DNA extraction, and 3) protocols for bone decalcifi­
cation and DNA extraction. Bone decalcification and various DNA extraction protocols have been tested and
optimised for ancient bone, suggesting that prolonged EDTA (Ethylenediaminetetraacetic acid) demineralisation
followed by solid-phased silica-based extraction techniques provide the greatest DNA yield. However, there is
significantly less literature exploring the optimal protocol for incinerated bones. Although burned bone, like
ancient and diagenetic bone, can be considered “low-copy”, the taphonomic processes occurring are likely
different. As techniques developed for ancient samples are tailored to deal with bone that has been altered in a
particular way, it is important to understand if burned bone undergoes similar or different changes. Currently the
effects of burning on bone and the DNA within it is not fully understood. Future research should focus on
increasing our understanding of the effects of heat on bone and on comparing the outcome of various DNA
extraction protocols for these tissues.

1. Introduction
Previous research on DNA retrieval from thermally damaged tissues
has been primarily palaeontology based, with a focus on establishing
phylogeny of ancient remains. Far less attention has been given to
retrieval of genetic information from incinerated tissues for forensic
human identification of modern samples, for example identification of
disaster victims. Twenty years ago, geneticists were not included in
disaster victim identification mortuary teams, as dental records were
considered a more effective means of establishing identity [1,2]. More
recently the prevalence of DNA analysis in forensic identification has
increased, although the process can still be costly and time consuming.
Even in contemporary remains, tissues may be compromised by
factors such as fragmentation, burning, comingling of remains, and
prolonged burial [3]. Some of these factors only damage the structural
integrity of the tissue while viable DNA may still be attainable [4].
However, in the event of extreme tissue damage (eg. casualty events
such as bush fires, structural fires, or bombings) even the DNA in skeletal
elements can become contaminated or degraded. Following the
September 11, 2001 World Trade Centre attack, DNA analysis was
considered invaluable for identification as remains were fragmented,
making identification from dentition alone almost impossible [5,6].
Similarly, during the 2009 Victoria bushfires forensic scientists took
post-mortem DNA samples and compared them to ante-mortem samples
from living individuals with missing relatives [7]. To obtain a usable
DNA profile in cases like these there are many considerations involved in
the process (Fig. 1) for example: which bones should be prioritised for
maximum DNA yield, what DNA extraction process should be adopted to
minimise contamination and inhibition, and what technique should be
used for downstream analysis [8].
Over the past few decades, the methods used to obtain genetic pro­
files for identification of casualties of war and missing persons have

* Corresponding author.
E-mail address: meghan.mckinnon@adelaide.edu.au (M. Mckinnon).

https://doi.org/10.1016/j.scijus.2021.03.006
Received 17 December 2020; Received in revised form 16 February 2021; Accepted 13 March 2021
Available online 23 March 2021
1355-0306/© 2021 The Chartered Society of Forensic Sciences. Published by Elsevier B.V. All rights reserved.

31
M. Mckinnon et al.
improved dramatically [9–12]. With modern techniques providing more
accurate results and becoming less labour intensive and less time
consuming the ability to successfully investigate degraded and
compromised remains genetically is becoming more likely. However,
skeletal remains affected by incineration remain a challenge. An
increased understanding of the changes that occur in burned bone and
the potential implications to successful genetic analysis of these tissues
will increase the potential to successfully identify individuals even when
the remains are severely damaged. This paper aims to collate all
currently available information with regards to the changes that occur in
burned bone, the potential implications these changes have to successful
genetic analysis, and current techniques and their limitations for this
sample type.
2. Sample collection
2.1. Bone diagenesis
To improve the likelihood of success the practitioner must select
which skeletal element to sample based on likely DNA yield. It has been
suggested that bone density influences DNA yield, as denser tissues
provide greater physical protection against damage [13,14]. For this
reason, DNA is often sampled from teeth, which are protected by hard
enamel [15–17], or dense, weight-bearing long bones (tibia or femur)
[18,19]. Although some studies have suggested small skeletal elements
of the distal skeleton (tarsal and carpal bones) can provide higher
quantities of DNA than other skeletal elements [20–22], this is unlikely
to be the case with incinerated remains. Bones of the hands and feet are
often completely destroyed or lost during incineration due to their small
size and the lack of soft tissues and density that affords significant
protection to larger bones [23]. Rapid moisture loss during exposure to
heat causes contraction of flexor muscles, which results in the charac­
teristic pugilistic “boxers” pose seen in incineration victims [23]. This
contraction also results in subsequent fracturing of the distal meta­
carpals and proximal phalanges, meaning these bones are rapidly
affected by incineration and less than optimal for DNA sampling. The
petrous (or petrosal) portion of the temporal bone has a greater density
than many other skeletal sites and has been shown to yield much greater
quantities of useable DNA, in some cases between four and 16 times
more than teeth [24–28]. Use of the petrous bone as a source of DNA has
largely been restricted to ancient remains, and potential uses in forensic
cases and with incinerated remains have not yet been fully investigated.
Research comparing DNA quality and quantity between different
Fig. 1. Flowchart visualising the process of DNA recovery. Figure created with BioRender.com.
Science & Justice 61 (2021) 332–338
skeletal elements (e.g. long bone compared to petrous bone) under the
influence of incineration would yield a better understanding of which
bones should be prioritised for increased efficiency of DNA sampling in
these cases.
Although this paper focuses on the effects of burning, it is important
to note that in forensic case work skeletal samples are often exposed to
multiple taphonomic effects, sometimes concurrently. Following (or
prior to) incineration bones may also be subjected to various other forms
of destruction and degradation, including animal predation, weathering
and burial. DNA is preserved differentially across different skeletal el­
ements- in fact differential DNA yield has even been found between
different regions of the same bone (i.e. low yield femoral diaphysis vs.
high yield epiphyses) [29], although the reasons behind these differ­
ences are not yet fully understood. Bones exposed to the elements can be
subject to diagenesis which Collins et al. [30] break down into three
pathways: 1) chemical degradation of organic bone material, 2) chem­
ical deterioration of bone minerals and 3) invasion of microbes. These
processes both increase the likelihood of contamination with exogenous
DNA and environmental contaminants and decrease the organic content
of the bone, resulting in a lower yield of viable DNA [30,31]. Weathering
can be defined as chemical and mechanical deterioration and destruc­
tion occurring over time, the rate and severity of which is influenced by
factors such as bone size, species and burial conditions [32]. Whether
there is a direct correlation between bone appearance and likely success
of genetic analysis remains debated. Misner et al. [33] investigated the
correlation between the physical appearance of weathered bone and
subsequent mitochondrial DNA quality and quantity. The authors
concluded that the degree of visible bone weathering, although likely to
reduce sample availability, had little effect on the DNA with no signif­
icant difference between differentially preserved samples.
Many studies have linked bone colouration and texture following
fires to different burn durations and temperatures [34–42]. Walker et al.
[43] used imaging software to generate average colour values for bone
samples subjected to selected temperatures and durations of fire and
concluded that collagen content could be estimated from bone colour.
Fredericks et al. [44] compared bovine DNA yield following differential
burning to bone colour using handheld digital colorimeters. This study
confirmed that bone colouration could be linked to the successful (or
unsuccessful) amplification of DNA. However, assessing bone colour­
ation is a) somewhat subjective, and b) beholden to other variables
besides heat. Additionally, uncontrolled burning, as occurs in disasters
and industrial or domestic events does not occur at a constant temper­
ature and hence will not necessarily mimic trial situations.
32
M. Mckinnon et al.
2.2. Microscopic and macroscopic bone analysis
Prior to subjecting a sample to DNA extraction, the likelihood of
success may be able to be assessed to some degree using microscopic and
macroscopic observations. As presence of nucleated cells and state of
tissue osteons can be directly linked to DNA content quantifying these
factors can give some indication of the likelihood of obtaining viable
endogenous DNA [15,45–48]. Castillo et al. [49] assessed changes to the
histological structure of bone following burning and found that at
300–400 ◦ C all organic material disappeared, and crystalline structures
began to form. Brain [50] described similar cellular changes in bone
following differential burning. Although Brain did not investigate the
presence of nucleated cells, the breakdown of other structures
mentioned (carbon accumulation in lacunae, cracks appearing in the
bone matrix, fusion of hydroxyapatite crystals) must also be considered,
as they are likely indicators of DNA degradation.
In particular, the formation of hydroxyapatite crystals following
burning may have implications for DNA extraction – if crystal aggregates
in modern burned bones form in the same way as in diagenetically
altered ancient bone, demineralisation techniques designed to extract
endogenous DNA from these tightly-bound aggregates could be useful in
forensic cases. This has been investigated to a degree using a combina­
tion of X-ray diffraction (XRD) and scanning electron microscopy (SEM)
to compare average size of bone hydroxyapatite crystals between
differentially preserved samples. The average size of hydroxyapatite
crystals in human and bovine bone and teeth has been shown to increase
exponentially at temperature increments above 500 ◦ C [51].
Bone collagen content is another factor often considered directly
linked to DNA yield, particularly in ancient remains [52]. Collagen
diagenesis within ancient bone has been studied, however the mecha­
nisms by which this process improves DNA survivability are somewhat
unclear. One theory is that as collagen becomes mineralised, DNA binds
to the positively charged hydroxyapatite calcium ions, which stabilises
DNA and protects it from degradation [53,54]. Others have stated that
mummification of individual osteocytes protected from contaminants by
the small size of bone pores determines survivability of DNA in bone
[55]. In regard to incinerated bone, studies have shown collagen is de­
natured at around 155 ◦ C, and all traces of organic material are lost by
approximately 400 ◦ C [56,57]. However, the possible correlation be­
tween heat-induced collagen degradation and DNA loss has not been
investigated. This is inherently difficult to inspect as DNA extraction
techniques are not designed to target DNA within specific individual
elements of the bone (e.g. hydroxyapatite crystals, mummified osteo­
cytes), but rather all DNA within a sample. In forensic cases, it may be
more beneficial to establish a means of assessing likely DNA content
from gross morphology (e.g. bone colour, collagen content) to estimate
the likelihood of collecting a usable sample. Theoretically, the same
principle could be applied to hydroxyapatite crystal size using simple
techniques such as SEM or XRD (X-Ray Diffraction). To do this the
correlation between burn temperature, hydroxyapatite crystal size and
DNA yield would have to be investigated.
3. Sample preparation
3.1. Tissue sampling
When working with incinerated remains samples cannot simply be
cut from the bones’ surface and used in the same way as an uncom­
promised soft tissue sample but must be processed in a manner which
reduces contaminants whilst maintaining sufficient material for DNA
extraction [58]. Finding viable sampling sites is especially difficult in
burned remains as the loss of organic bone components at high tem­
peratures can decrease bone size by 20% [59–61], rendering the
remaining material warped and shrunken [34]. During extraction,
improper handling and sample storage can lead to cross-contamination
of target DNA with non-target DNA, which is often the case in mass
Science & Justice 61 (2021) 332–338
disaster situations where inexperienced personnel are involved in sam­
ple recovery [62–64]. Additionally, environmental inhibitors such as
bacteria, fulvic and humic acids, tannins, phenolic compounds, and ions
can result in failure of downstream applications [65]. This can be
avoided by thoroughly cleaning the bones’ surface before extraction,
after which all downstream processes should be performed in a dedi­
cated laminar flow cabinet using UV-treated tools [66].
Cleaning commonly involves physical removal of the surface layer
via sanding or drilling, followed by gentle rubbing with ethanol and/or
sodium hypochlorite (NaOCl) to destroy DNA on the newly exposed
bone surface [66,67]. Using this method, charring to the bones’ exterior
can sometimes be removed to expose less affected surfaces. However, it
is often unclear how deep burning has penetrated the bone, and if the
sample is entirely calcined the bone may be too fragile to survive this
process. A clean sample must be pulverized to maximize surface area to
volume ratio for successful release of DNA during extraction [68].
Collection of powdered bone may be achieved by drilling provided no
heat is generated [69] as heat can further degrade any surviving DNA.
The use of NaOCl, although effective for removal of DNA contamination,
can also destroy targeted endogenous DNA. Korlević et al. [70] found
that NaOCl treated samples contained significantly less DNA contami­
nation, however 63% of the target DNA was also lost in the process.
Furthermore, Dissing et al. [71] investigated the migration of ClO- ions
from decontaminating agents into the tooth pulp cavity, and the sub­
sequent effects on DNA yield using a 2% NaOCl solution to soak teeth for
different durations (5 min and 30 min respectively). Results showed
ClO- ions did indeed migrate into the tooth interior when applied to the
enamel surface, but authentic DNA was still recoverable and exogenous
DNA successfully removed if the duration of NaOCl exposure was
limited to 5 min. Although not thoroughly tested using bone tissue, due
to the increased porosity of its internal structure, migration of oxidizing
ions into the bone matrix would likely be increased resulting in a greater
loss of target DNA. Thus, submerging bone in NaOCl for prolonged du­
rations may not be the best technique for cleaning these types of sam­
ples, and simply rubbing the surface of the bone with NaOCl followed by
immediate drying may provide better results.
3.2. Sample demineralisation
Demineralisation involves removing the inorganic component of the
bone to liberate DNA containing cells bound or trapped in the mineral
matrix [15,72]. If demineralisation is not complete unreleased DNA may
be discarded during filtering and if calcium ions are co-extracted with
target material they may bind to target DNA and/or necessary reagents
and inhibit polymerase chain reaction (PCR) [73]. Tissue is routinely
decalcified using acidic agents such as hydrochloric acid (HCl), RDO
Gold or Ethylenediaminetetraacetic acid (EDTA) [47,74]. Choi et al.
[75] compared the outcome of demineralisation using the aforemen­
tioned agents for histomorphologic and DNA analyses, and found EDTA
best preserved genetic material, a claim supported by many similar
studies [46,47,76,77]. The method of demineralisation may require
modification depending on the specific sample. In cases involving
degraded material, total demineralisation has proven the most effective
technique for achieving high DNA yields from bone and tooth samples
[72,78–80]. Total pulverization (e.g. via cryogenic grinding) has shown
to increase the rate of demineralisation and digestion of bone [81]. The
advantage of total sample pulverization has been demonstrated in many
studies; Rohland & Hofreiter [82] found the fine powder samples yiel­
ded significantly more DNA. Similarly, Sweet and Hildebrand [83]
found cryogenically pulverising teeth to access the DNA-rich centre was
faster and more effective than other methods of sampling and signifi­
cantly reduced contamination.
Demineralisation is critical for ancient bone, as these tissue types are
extremely petrified, and the breakdown of hydroxyapatite takes much
longer than in fresh tissues [84] – it is not known if this is also the case
for incinerated bones. Much like burned bones, ancient samples contain
33
M. Mckinnon et al.
little organic material to house DNA, and what remains is bound to
dense crystal aggregates of hydroxyapatite [71]. Although liberating
DNA from these aggregates can be challenging, the crystalline structure
is extremely resistant to degradation and contamination, providing
excellent protection to the DNA even from common oxidizing agents
such as NaOCl and inhibitory exogenous material [85].
The process of total demineralisation involves submerging the skel­
etal sample in a chelating solution such as EDTA to remove the miner­
alised component before incubation in a lysis buffer. In some cases,
multiple EDTA washes have been used to encourage complete dissolu­
tion of the bone. Discarding the wash solutions is considered to reduce
contaminants [86,87]. However, Loreille et al. [72] hypothesised that in
some circumstances discarding EDTA supernatant resulted in DNA loss.
A solution to this was to integrate EDTA into the lysis buffer, eliminating
the EDTA washing step and maintaining reagent volume and concen­
tration, as well as providing 4.6 times more DNA on average than older
protocols. Loreille’s total demineralisation protocol is generally
considered one of the most effective methods of DNA extraction from
challenging bone samples. With regards to incinerated bone, however, a
recent study [88] comparing the effectiveness of this protocol to the
incomplete decalcification protocol described by Dabney [89], found
that the later improved recovery of shorter DNA molecules (<185 bp).
Although most research agrees that ancient/diagenetic bones require
prolonged, total demineralisation, notably less information exists
regarding treatment of burned bones. A study by Ye et al. [58] indicated
that as with ancient bones, an adapted EDTA extraction protocol could
improve DNA extraction from charred bones. A major adaptation to the
standard extraction protocol was the duration of EDTA demineralisa­
tion. The length of time required to completely demineralise a sample
can be a trade-off between maximising the quantity of DNA extracted
over the throughput of samples [90]. In cases where samples are few and
precious (for example ancient bone), an over-conservative approach is
justified. This is not always the case in forensic and disaster victim
identification situations, where fast results are often required for a large
number of samples [4]. In these cases, demineralisation protocols may
require re-analysis. Remains subject to extreme trauma, for example
burning, are likely compromised in a similar way to ancient bone, with
decreased organic components although this has not been thoroughly
investigated [34,91].
4. DNA extraction
Following demineralisation DNA extraction is required. A majority
of research concerning burned bone DNA extraction is centred around
collection and demineralisation of samples. Far less attention has been
focused on subsequent extraction processes. All extraction techniques
follow the same basic process: DNA is separated from cellular debris,
isolated and purified. This can be achieved using phenol/chloroform,
whereby the sample is suspended in a mixture containing Dextran Blue,
acetate and EtOH, however due to the use of toxic chemicals (and
incomplete inhibitor removal), this method has become obsolete for
many samples in forensic casework [92]. A different class of extractions
is solid-phase DNA extraction, most of which use silica-based binding
that can double the amount of useable DNA recovered [93]. Unlike
organic methods, silica-based techniques rely on a simple bind, wash,
elute process [94]. The DNA is denatured, then binds to silica particles
that can be separated from inhibitors and non-target material [95,96].
This method is quicker, easier, more cost efficient and better for removal
of DNA inhibitors than other methods, therefore is especially useful in
low-copy DNA contexts [68]. The advantages of silica-based methods
has been thoroughly examined for ancient and modern degraded sam­
ples. A study by Pajnič et al. [97] presented an optimised silica-based
protocol for extracting DNA from challenging ancient and contempo­
rary remains. Although two burned femurs were included in this study,
it could be expanded with the inclusion of a series of incinerated bones
burned under comparable conditions. As it stands, there are no well-
Science & Justice 61 (2021) 332–338
established, routine methods of extracting DNA from incinerated mod­
ern bone.
There are two general approaches to silica-based extraction; sus­
pension in a silica solution and kit-based binding to silica beads or
membranes. Silica suspension is a common alternative to silica binding
that allows for processing of substantially larger sample volumes than
commercial kits [82]. This is a necessity for aDNA extraction where
large sample volumes containing small quantities of DNA are processed
in batches. However, Dabney et al [89] states that the efficiency of this
method is dependent on fragment length, with a strong bias towards
longer fragments. In situations where only short fragments are available
(as is often the case in forensic casework) using commercially available
kits in place of silica suspension may provide better results. These kit-
based methods are also preferred by forensic professionals as they are
simple to set up, easily automated, and validated for use. Two of the
main silica-binding methods are described in Qiagen kit user manuals
[98,99] which are presented in Fig. 2.
Rothe and Nagy [65] compared these two techniques and outlined
the advantages and disadvantages of both. Although DNA yield was
significantly greater using the membrane method compared to the
beads, co-extraction of inhibitors was also a much greater problem. Wen
et al. [100] suggests that capillary-based monoliths may be more
effective again than commercial spin column techniques with an effi­
ciency of 86% compared to 49% respectively. However, this was not the
case for larger sample volumes, and the lack of device stability and
reproducibility suggest that commercial technologies such as spin col­
umns are the best option at present [100,101]. When working with
ancient DNA demineralised EDTA supernatant is generally discarded
prior to extraction due to the high likelihood of co-extracting contami­
nation and inhibitors [72,86,87,97]. However, it is not unlikely that this
could vary in burned samples. Ancient bone contains virtually no
exogenous DNA and there is very little risk of discarding targeted ma­
terial in the supernatant. Comparatively, burned samples may still
contain exogenous DNA in addition to crystallite-bound endogenous
material, therefore co-extracting demineralised supernatant may actu­
ally improve DNA yield. Although this could increase the likelihood of
contamination, it is a potential trade-off that should be more thoroughly
investigated in future research.
5. Areas for expansion
Extracting DNA from (relatively) fresh modern bone tissue is a well-
established process; large companies supplying DNA equipment and
consumables (e.g. Qiagen) will generally provide clear protocols for
using these sample types. Furthermore, the protocol for extracting DNA
from ancient samples is well standardised (i.e. prolonged deminerali­
sation to liberate bound DNA). Interestingly no such protocol exists for
modern burned samples, and most decisions on DNA analysis in these
situations are made ad-hoc on a case-by-case basis. There is a growing
need for identification based on these sample types; for example, many
of the World Trade Centre remains were never identified due to DNA
contamination from commingling of remains [4]. Similarly, authorities
fear that many victims of the recent Grenfell tower fire in London may
never be identified as remains were almost completely incinerated and
DNA destroyed. As many residents were illegal immigrants, dental re­
cords were not available, and DNA was considered the only option for
identification [102].
Knowing that certain regions of the skeleton provide more protection
to DNA than others, research into differential DNA yield from non-
standard sampling sites following extreme tissue damage is essential.
If these non-standard sites (e.g. petrous bone, metacarpals) provide
higher quality and quantities of DNA, standard procedures for DNA
sampling during disaster victim identification and other mass casualty
events could be re-examined and adjusted if needed. A study by Kulstein
et al. [103] assessed different sites on the skeleton by their DNA content,
however this same analysis has not been conducted on incinerated
34
M. Mckinnon et al.
Fig. 2. Flowchart comparing silica-based methods of DNA extraction. Figure created with BioRender.com.
remains. If a graded system for DNA yield from incinerated bone was
established, this system could be used to prioritise sampling procedures
based on likely presence of DNA. Additionally, downstream DNA pro­
cesses could be refined to maximize DNA yield, minimize time used and
optimise accuracy of identifications. DNA analysis in human identifi­
cation has greatly advanced and is often considered the gold standard in
forensic science. Many studies have already presented strategies for
improving DNA sampling and extraction from degraded skeletal mate­
rial [104–107], however due to the differences between diagenetic and
incinerated bones, these techniques may not be applicable. This suggests
a need for further study applying some of these conservative extraction
techniques to incinerated samples.
Many conclusions made in the existing literature are formulated
from “best case scenario” research, using pristine bone samples prepared
in the laboratory under controlled conditions. However, this somewhat
defeats the purpose of such research – in a realistic situation, if bones are
well preserved and uncontaminated, there is usually equally well pre­
served soft tissue, which is much easier to sample for DNA analysis. In
this case, what is needed is a better understanding of DNA retrieval
under less than ideal circumstances, so the effect of these circumstances
(e.g. exposure to extreme heat) can be studied. A possible objective for
future work would be the establishment of standards to estimate likely
DNA yield based on simple tests such as ATR-FTIR spectroscopy, x-ray
diffraction and scanning electron microscopy [108]. This could also be
addressed by investigating whether the general macroscopic appearance
(e.g. colour, texture, size) of incinerated bones can be indicative of DNA
Science & Justice 61 (2021) 332–338
survivability, and whether various skeletal sites provide different
quantities of DNA. It is apparent that a multi-dimensional, multi-disci­
plinary approach is required to understand DNA on different levels – not
just DNA extraction and amplification, or histological analysis or gross
anatomy, but rather a combination of these techniques to establish the
most efficient and reliable way of sampling DNA from burned remains.
Funding
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
CRediT authorship contribution statement
Meghan Mckinnon: Conceptualization, Formal analysis, Investiga­
tion, Writing - original draft. Maciej Henneberg: Conceptualization,
Writing - review & editing, Supervision, Project administration, Funding
acquisition. Denice Higgins: Conceptualization, Investigation, Writing -
review & editing, Supervision, Project administration, Funding
acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
35
M. Mckinnon et al. Science & Justice 61 (2021) 332–338

References [31] A. Westen, R. Gerretsen, G. Maat, Femur, rib, and tooth sample collection for DNA
analysis in disaster victim identification (DVI), Forensic Sci. Med. Pathol. 4 (1)
(2008) 15–21.
[1] I. Pretty, D. Sweet, A look at forensic dentistry–Part 1: the role of teeth in the
[32] R. Lyman, G. Fox, A critical evaluation of bone weathering as an indication of
determination of human identity, Br. Dent. J. 190 (7) (2001) 359–366.
bone assemblage formation, J. Archaeol. Sci. 16 (3) (1989) 293–317.
[2] M. Petju, A. Suteerayongprasert, R. Thongpud, K. Hassiri, Importance of dental
[33] L. Misner, A. Halvorson, J. Dreier, D. Ubelaker, D. Foran, The correlation between
records for victim identification following the Indian Ocean tsunami disaster in
skeletal weathering and DNA quality and quantity, J. Forensic Sci. 54 (4) (2009)
Thailand, Public Health 121 (4) (2007) 251–257.
822–828.
[3] A. Mundorff, R. Shaler, E. Bieschke, E. Mar-Cash, Marrying anthropology and
[34] D. Dirkmaat, G. Olson, A. Klales, S. Getz, The role of forensic anthropology in the
DNA: essential for solving complex commingling problems in cases of extreme
recovery and interpretation of the fatal-fire victim, in: D. Dirkmaat (Ed.),
fragmentation, in: B. Adams, J. Byrd (Eds.), Commingled Human Remains,
A Companion to Forensic Anthropology, Wiley-Blackwell, 2012, pp. 113–135.
Academic Press, San Diego, 2014, pp. 257–273.
[35] T. Schwark, A. Heinrich, A. Preuße-Prange, Von Wurmb-Schwark, N 2011
[4] M. Holland, C. Cave, C. Holland, T. Bille, Development of a quality, high
‘Reliable genetic identification of burnt human remains’, Forensic Sci. Int. Genet.
throughput DNA analysis procedure for skeletal samples to assist with the
5 (5) (2011) 393–399.
identification of victims from the World Trade Center attacks, Croatian Med. J. 44
[36] Devlin, J & Herrmann, N 2008, ‘Bone colour as an interpretive tool of the
(3) (2003) 264–272.
depositional history of archaeological remains’, In: Schmit, C & Symes, S (eds),
[5] A. Mundorff, E. Bartelink, E. Mar-Cash, DNA preservation in skeletal elements
The Analysis of Burned Human Remains, pp. 190-128, Academic Press, London.
from the World Trade Center disaster: recommendations for mass fatality
[37] M. Hanson, C. Cain, Examining histology to identify burned bone, J. Archaeol.
management, J. Forensic Sci. 54 (4) (2009) 739–745.
Sci. 34 (2007) 1902–1913.
[6] G. MacKinnon, A. Mundorff, The World Trade Center—September 11, 2001, An
[38] P. Correia, Fire modification of bone: a review of the literature, in:
Introduction, Forensic Human Identification, 2006, pp. 485–499.
Forensic Taphonomy (Ed.), Haglund, W, Sorg, M, CRC Press, New York, 1997,
[7] D. Hartman, O. Drummer, C. Eckhoff, J. Scheffer, P. Stringer, The contribution of
pp. 275–293.
DNA to the disaster victim identification (DVI) effort, Forensic Sci. Int. 205 (1–3)
[39] P. McCutcheon, Burned archaeological bone, in: J. Stein (Ed.), Deciphering a
(2011) 52–58.
Shell Midden, Academic Press, San Diego, 1992, pp. 347–368.
[8] L. Biesecker, J. Bailey-Wilson, J. Ballantyne, H. Baum, F. Bieber, C. Brenner, et
[40] P. Shipman, G. Foster, M. Schoeninger, Burnt bones and teeth: an experimental
al., DNA identifications after the 9/11 world trade center attack, Science 310
study of color, morphology, crystal structure and shrinkage, J. Archaeol. Sci. 11
(5751) (2005) 1122–1123.
(1984) 307–325.
[9] Jobling, M & Gill 2004, ‘Encoded evidence: DNA in forensic analysis’, Nature
[41] C. Brain, The Hunters or the Hunted? An Introduction to African Cave
Reviews Genetics, vol. 5, no. 10, pp. 739.
Taphonomy, University of Chicago Press, Chicago, 1981.
[10] I. Pajnič, M. Obal, T. Zupanc, Identifying victims of the largest Second World War
[42] E. Bonucci, G. Graziani, ‘Comparative thermogravimetric, x-ray diffraction and
family massacre in Slovenia, Forensic Sci. Int. 306 (2020), 110056.
electron microscope investigations of burnt bone from recent, ancient and
[11] I. Morild, S. Hamre, R. Huel, T. Parsons, Identification of missing norwegian
prehistoric age’, Atti Accademia Nazionale dei Lincei, Classe di Scienze, Fisiche,
world war II soldiers, in Karelia Russia, J. Forensic Sci. 60 (4) (2015) 1104–1110.
Matematiche e Naturali 59 (1975) 517–532.
[12] T. Parsons, R. Huel, Z. Bajunović, A. Rizvić, Large scale DNA identification: The
[43] P. Walker, K. Miller, M. Richman, Time, temperature, and oxygen availability: an
ICMP experience, Forensic Sci. Int. Genet. 38 (2019) 236–244.
experimental study of the effects of environmental conditions on the color and
[13] Galloway, A, Willy, P, Snyder, L., 1997, ’Human bone mineral densities and
organic content of cremated bone, in: C.W. Schmidt, S.A. Symes (Eds.), The
survival of bone elements: a contemporary sample.’, in SM Haglund WD (ed.),
analysis of burned human remains, Academic Press, 2008, pp. 129–135.
Forensic taphonomy: the postmortem fate of human remains., Boca Ration.
[44] J. Fredericks, T. Ringrose, A. Dicken, A. Williams, P. Bennett, A potential new
[14] M. Leney, Sampling skeletal remains for ancient DNA (aDNA): a measure of
diagnostic tool to aid DNA analysis from heat compromised bone using
success, Histor. Archaeol. 40 (3) (2006) 31–49.
colorimetry: a preliminary study, Sci. Justice 55 (2015) 124–130.
[15] D. Higgins, A.B. Rohrlach, J. Kaidonis, G. Townsend, J. Austin, Differential
[45] J. Edson, E. Brooks, C. McLaren, J. Robertson, D. McNevin, A. Cooper, et al.,
nuclear and mitochondrial DNA preservation in post-mortem teeth with
A quantitative assessment of a reliable screening technique for the STR analysis of
implications for forensic and ancient DNA studies, PLoS ONE 10 (5) (2015) 1–17.
telogen hair roots, Forensic Sci. Int. Genet. 7 (1) (2013) 180–188.
[16] P. Malaver, J. Yunis, Different dental tissues as source of DNA for human
[46] V. Singh, R. Salunga, V. Huang, Y. Tran, M. Erlander, P. Plumlee, et al., Analysis
identification in forensic cases, Croat. Med. J. 44 (3) (2003) 306–309.
of the effect of various decalcification agents on the quantity and quality of
[17] R. Gaytmenn, D. Sweet, Quantification of forensic DNA from various regions of
nucleic acid (DNA and RNA) recovered from bone biopsies, Ann. Diagnost.
human teeth, J. Forensic Sci. 48 (3) (2003) 622–625.
Pathol. 17 (4) (2013) 322–326.
[18] D. Hines, M. Vennemeyer, S. Amory, R. Huel, I. Hanson, C. Katzmarzyk, et al.,
[47] J. Alers, P. Krijtenburg, K. Vissers, H. Van Dekken, Effect of bone decalcification
Prioritized sampling of bone and teeth for DNA analysis in commingled cases,
procedures on DNA In situ hybridization and comparative genomic hybridization:
Coming. Hum. Remains (2014) 275–305.
EDTA is highly preferable to a routinely used acid decalcifier, J. Histochem.
[19] A. Milos, A. Selmanovic, L. Smajlovic, R. Huel, C. Katzmarzyk, A. Rizvic, et al.,
Cytochem. 47 (5) (1999) 703–709.
Success rates of nuclear short tandem repeat typing from different skeletal
[48] C. Cattaneo, S. Dimartino, S. Scali, O. Craig, M. Grandi, R. Sokol, Determining the
elements, Croat. Med. J. 48 (4) (2007) 486–493.
human origin of fragments of burnt bone: a comparative study of histological,
[20] T. Zupanc, E. Podovšovnik, M. Obal, I. Pajnič, High DNA yield from metatarsal
immunological and DNA techniques, Forensic Sci. Int. 102 (2–3) (1999) 181–191.
and metacarpal bones from Slovenian Second World War skeletal remains,
[49] F. Castillo, D. Ubelaker, J. Acosta, R. Rosa, I. Garcia, Effect of temperature on
Forensic Sci. Int. Genet. 51 (2021), 102426.
bone tissue: histological changes, J. Forensic Sci. 58 (3) (2013) 578–582.
[21] J. Watherston, D. McNevin, M. Gahan, D. Bruce, J. Ward, Current and emerging
[50] C. Brain, The occurrence of burnt bones at Swartkrans and their implications for
tools for the recovery of genetic information from post mortem samples: new
the control of fire by early hominids, in: C. Brain (Ed.), Swartkrans: A Cave’s
directions for disaster victim identification, Forensic Sci. Int. Genet. 37 (2018)
Chronicle of Early Man, Transvaal Museum, Pretoria, 1993, pp. 229–242.
270–282.
[51] G. Piga, A. Santos-Cubedo, S. Solà, A. Brunetti, A. Malgosa, S. Enzo, An X-ray
[22] A. Mundorff, J. Davoren, Examination of DNA yield rates for different skeletal
diffraction (XRD) and X-ray fluorescence (XRF) investigation in human and
elements at increasing post mortem intervals, Forensic Sci. Int. Genet. 8 (1)
animal fossil bones from Holocene to Middle Triassic, J. Archaeol. Sci. 36 (9)
(2014) 55–63.
(2009) 1857–1868.
[23] S. Symes, C. Rainwater, E. Chapman, D. Gipson, A. Piper, Patterned thermal
[52] P. Campos, O. Craig, G. Turner-Walker, E. Peacock, E. Willerslev, M. Gilbert, DNA
destruction of human remains in a forensic setting, in: The analysis of burned
in ancient bone–Where is it located and how should we extract it? Ann. Anatom.
human remains, Academic Press, 2008, pp. 15–vi.
Anatom. Anzeiger 194 (2011) 7–16.
[24] H. Hansen, P. Damgaard, A. Margaryan, J. Stenderup, N. Lynnerup, E. Willerslev,
[53] A. Götherström, M. Collins, A. Angerbjörn, K. Lidén, Bone preservation and DNA
et al., Comparing ancient DNA preservation in petrous bone and tooth cementum,
amplification, Archaeometry 44 (3) (2002) 395–404.
PLoS ONE 12 (1) (2017) 1–18.
[54] M. Collins, M. Riley, A. Child, G. Turner-Walker, A basic mathematical simulation
[25] R. Pinhasi, D. Fernandes, K. Sirak, M. Novak, S. Connell, S. Alpaslan-Roodenberg,
of the chemical degradation of ancient collagen, J. Archaeol. Sci. 22 (2) (1995)
et al., Optimal ancient DNA yields from the inner ear part of the human petrous
175–183.
bone, PLoS ONE 10 (6) (2015), e0129102.
[55] C. Lassen, S. Hummel, B. Herrmann, Comparison of DNA extraction and
[26] C. Gamba, E. Jones, M. Teasdale, R. McLaughlin, G. Gonzalez-Fortes,
amplification from ancient human bone and mummified soft tissue, Int. J. Legal
V. Mattiangeli, et al., Genome flux and stasis in a five millennium transect of
Med. 107 (3) (1994) 152–155.
European prehistory, Nat. Commun. 5 (2014) 1–9.
[56] M. Harbeck, R. Schleuder, J. Schneider, I. Wiechmann, W. Schmahl, G. Grupe,
[27] A. Gonzalez, C. Cannet, V. Zvénigorosky, A. Geraut, G. Koch, T. Delabarde, et al.,
Research potential and limitations of trace analyses of cremated remains,
The petrous bone: Ideal substrate in legal medicine? Forensic Sci. Int. Genet. 47
Forensic Sci. Int. 204 (1–3) (2011) 191–200.
(2020), 102305.
[57] N. Kalsbeek, J. Richter, Preservation of burned bones: an investigation of the
[28] E. Pilli, S. Vai, M. Caruso, G. D’Errico, A. Berti, D. Caramelli, Neither femur nor
effects of temperature and pH on hardness, Stud. Conserv. 51 (2) (2006)
tooth: petrous bone for identifying archaeological bone samples via forensic
123–138.
approach, Forensic Sci. Int. 283 (2018) 144–149.
[58] J. Ye, A. Ji, E. Parra, X. Zheng, C. Jiang, X. Zhao, et al., A simple and efficient
[29] C. Antinick, R. Foran, Intra-and inter-element variability in mitochondrial and
method for extracting DNA from old and burned bone, J. Forensic Sci. 49 (4)
nuclear DNA from fresh and environmentally exposed skeletal remains,
(2004) 1–6.
J. Forensic Sci. 64 (1) (2019) 88–97.
[59] Herrmann, B 1977, ‘On histological investigations of cremated human remains’,
[30] Collins, M, Nielsen–Marsh, C, Hiller, J, Smith, C, Roberts, J, Prigodich, R, et al.
Journal of Human Evolution, vol. 6, pp. 2, pp. 101.
2002, ‘The survival of organic matter in bone: a review’, Archaeometry, vol. 44,
no. 3, pp. 383-394.

36
M. Mckinnon et al.
[60] P. Mayne-Correia, O. Beattie, A critical look at methods for recovering,
evaluating, and interpreting cremated human remains, in: W. Haglund, M. Sorg
(Eds.), Advances in Forensic Taphonomy: Method, Theory, and Archaeological
Perspectives, CRC Press, Boca Raton, 2002, pp. 435–450.
[61] T. Thompson, Heat-induced dimensional changes in bone and their consequences
for forensic anthropology, J. Forensic Sci. 50 (5) (2005) 1008–1015.
[62] N. von Wurmb-Schwark, A. Heinrich, M. Freudenberg, M. Gebühr, T. Schwark,
The impact of DNA contamination of bone samples in forensic case analysis and
anthropological research, Leg. Med. 10 (3) (2008) 125–130.
[63] A. Alonso, P. Martin, C. Albarran, P. Garcia, L. Fernandez de Simon, M. Jesus, et
al., Challenges of DNA profiling in mass disaster investigations, Croat. Med. J. 46
(4) (2005) 540–548.
[64] B. Budowle, F. Bieber, A. Eisenberg, Forensic aspects of mass disasters: Strategic
considerations for DNA-based human identification, Leg. Med. 7 (4) (2005)
230–243.
[65] J. Rothe, M. Nagy, Comparison of two silica-based extraction methods for DNA
isolation from bones, Leg. Med. 22 (2016) 36–41.
[66] R. Li, L. Liriano, A bone sample cleaning method using trypsin for the isolation of
DNA, Leg. Med. 13 (6) (2011) 304–308.
[67] B. Kemp, D. Smith, Use of bleach to eliminate contaminating DNA from the
surface of bones and teeth, Forensic Sci. Int. 154 (1) (2005) 53–61.
[68] N. Rohland, M. Hofreiter, Ancient DNA extraction from bones and teeth, Nat.
Protoc. 2 (2007) 1756–1762.
[69] Adler, C, Haak, W, Donlon, D, Cooper, A & Consortium, G 2011, ‘Survival and
recovery of DNA from ancient teeth and bones’, Journal of Archaeological
Science, vol. 38, pp. 956-964.
[70] P. Korlević, T. Gerber, M. Gansauge, M. Hajdinjak, S. Nagel, A. Aximu-Petri, et al.,
Reducing microbial and human contamination in DNA extractions from ancient
bones and teeth, Biotechniques 59 (2) (2015) 87–93.
[71] J. Dissing, M. Kristinsdottir, C. Friis, On the elimination of extraneous DNA in
fossil human teeth with hypochlorite, J. Archaeol. Sci. 35 (6) (2008) 1445–1452.
[72] O. Loreille, T. Diegoli, J. Irwin, M. Coble, T. Parsons, High efficiency DNA
extraction from bone by total demineralization, Forensic Sci. Int. Genet. 1 (2007)
191–195.
[73] K. Opel, D. Chung, B. McCord, A study of PCR inhibition mechanisms using real
time PCR, J. Forensic Sci. 55 (1) (2010) 25–33.
[74] D. Lahiri, B. Schnabel, DNA isolation by a rapid method from human blood
samples: effects of MgCl 2, EDTA, storage time, and temperature on DNA yield
and quality, Biochem. Genet. 31 (7) (1993) 321–328.
[75] S. Choi, S. Hong, S. Yoon, Proposal of an appropriate decalcification method of
bone marrow biopsy specimens in the era of expanding genetic molecular study,
J. Pathol. Trans. Med. 49 (3) (2015) 236–242.
[76] R. Brown, J. Edwards, J. Bartlett, C. Jones, A. Dogan, Routine acid decalcification
of bone marrow samples can preserve DNA for FISH and CGH studies in
metastatic prostate cancer, J. Histochem. Cytochem. 50 (1) (2002) 113–115.
[77] S. Tinling, R. Giberson, R. Kullar, Microwave exposure increases bone
demineralization rate independent of temperature, J. Microsc. 215 (3) (2004)
230–235.
[78] J. Jakubowska, A. Maciejewska, R. Pawlowski, Comparison of three methods of
DNA extraction from human bones with different degrees of degradation, Int. J.
Legal Med. 126 (2012) 173–178.
[79] D. Vanek, M. Silerova, V. Urbanova, L. Saskova, J. Dubska, M. Beran, Genomic
DNA extraction protocols for bone samples: the comparison of Qiagen and Zymo
Research spin columns, Foren. Sci. Int. Genet. Suppl. Ser. 3 (1) (2011) e397–e398.
[80] S. Seo, A. Zhang, H. Kim, J. Yi, H. Lee, D. Shin, et al., Technical note: Efficiency of
total demineralization and ion-exchange column for DNA extraction from bone,
Am. J. Phys. Anthropol. 141 (2010) 158–162.
[81] E. Morales Colón, M. Hernández, M. Candelario, M. Meléndez, T. Dawson Cruz,
Evaluation of a freezer mill for bone pulverization prior to DNA extraction: an
improved workflow for STR analysis, J. Forensic Sci. 63 (2) (2018) 530–535.
[82] N. Rohland, H. Siedel, M. Hofreiter, A rapid column-based ancient DNA
extraction method for increased sample throughput, Mol. Ecol. Resour. 10 (4)
(2010) 677–683.
[83] D. Sweet, D. Hildebrand, Recovery of DNA from human teeth by cryogenic
grinding, J. Forensic Sci. 43 (6) (1998) 1199–1202.
[84] C. Mulligan, Isolation and analysis of DNA from archaeological, clinical, and
natural history specimens, in: E. Zimmer, E. Roalson (Eds.), Methods in
enzymology, vol. 395, Academic Press, 2005, pp. 87–103.
Science & Justice 61 (2021) 332–338
[85] M. Salamon, N. Tuross, B. Arensburg, S. Weiner, Relatively well preserved DNA is
present in the crystal aggregates of fossil bones, Proc. Natl. Acad. Sci. 102 (39)
(2005) 13783–13788.
[86] A. Mohammadi, A. Ghorbani, M. Khafaei, et al., A new and efficient method for
DNA extraction from human skeletal remains usable in DNA typing, J. Appl.
Biotechnol. Rep. 4 (2017) 609–614.
[87] P. Damgaard, A. Margaryan, H. Schroeder, et al., Improving access to endogenous
DNA in ancient bones and teeth, Sci. Rep. 5 (2015) 11184.
[88] M. Emery, K. Bolhofner, S. Winingear, R. Oldt, M. Montes, S. Kanthaswamy, et al.,
Reconstructing full and partial STR profiles from severely burned human remains
using comparative ancient and forensic DNA extraction techniques, Forensic Sci.
Int. Genet. 46 (2020), 102272.
[89] J. Dabney, M. Knapp, I. Glocke, M. Gansauge, A. Weihmann, B. Nickel, et al.,
Complete mitochondrial genome sequence of a Middle Pleistocene cave bear
reconstructed from ultrashort DNA fragments, Proc. Natl. Acad. Sci. 110 (39)
(2013) 15758–15763.
[90] A. Cho, S. Suzuki, J. Hatakeyama, N. Haruyama, A. Kulkarni, A method for rapid
demineralization of teeth and bones, Open Dentistry J. 4 (2010) 223.
[91] M. Stiner, S. Kuhn, S. Weiner, O. Bar-Yosef, Differential burning, recrystallization,
and fragmentation of archaeological bone, J. Archaeol. Sci. 22 (2) (1995)
223–237.
[92] Herrmann, B 1977, ‘On histological investigations of cremated human remains’,
Journal of Human Evolution, vol. 6, pp. 2, pp. 101.
[93] J. Davoren, D. Vanek, R. Konjhodzić, J. Crews, E. Huffine, T. Parsons, Highly
effective DNA extraction method for nuclear short tandem repeat testing of
skeletal remains from mass graves, Croatian Med. J. 48 (4) (2007) 478.
[94] Tan, S & Yiap, B 2009, ‘DNA, RNA, and protein extraction: the past and the
present’, BioMed Research International, vol. 2009.
[95] M. Höss, S. Pääbo, DNA extraction from Pleistocene bones by a silica-based
purification method, Nucleic Acids Res. 21 (16) (1993) 3913.
[96] R. Boom, C. Sol, M. Salimans, C. Jansen, P. Wertheim-van Dillen, et al., Rapid and
simple method for purification of nucleic acids, J. Clin. Microbiol. 28 (3) (1990)
495–503.
[97] I. Pajnič, M. Debska, B. Pogorelc, K.V. Mohorčič, J. Balažic, T. Zupanc, et al.,
Highly efficient automated extraction of DNA from old and contemporary skeletal
remains, J. Forensic Leg. Med. 37 (2016) 78–86.
[98] Qiagen, 2008. MinElute® MinElute PCR Purification Kit Handbook. Qiagen:
Hilden Germany.
[99] Qiagen, 2014. EZ1® DNA Investigator® Handbook. Qiagen: Hilden Germany.
[100] J. Wen, C. Guillo, J. Ferrance, J. Landers, DNA extraction using a tetramethyl
orthosilicate-grafted photopolymerized monolithic solid phase, Anal. Chem. 78
(5) (2006) 1673–1681.
[101] K. Wolfe, M. Breadmore, J. Ferrance, M. Power, J. Conroy, P. Norris, et al.,
Toward a microchip-based solid-phase extraction method for isolation of nucleic
acids, Electrophoresis 23 (5) (2002) 727–733.
[102] ‘Grenfell Tower: New photos reveal ’indescribable’ scene as detectives hunt for
bodies in inferno apartment block’, ABC News, 19 June 2017, accessed 11 July
2017.
[103] G. Kulstein, T. Hadrys, P. Wiegand, As solid as a rock—comparison of CE-and
MPS-based analyses of the petrosal bone as a source of DNA for forensic
identification of challenging cranial bones, Int. J. Legal Med. 132 (1) (2018)
13–24.
[104] I. Pajnič, P. Fattorini, Strategy for STR typing of bones from the Second World
War combining CE and NGS technology: a pilot study, Forensic Sci. Int. Genet. 50
(2021), 102401.
[105] I. Pajnič, B. Pogorelc, T. Zupanc, Next generation sequencing technology in
Second World War victim identification, Forensic Sci. Int. Genet. Suppl. Ser. 7 (1)
(2019) 123–125.
[106] F. Calafell, R. Anglada, N. Bonet, M. González-Ruiz, G. Prats-Muñoz, R. Rasal, et
al., An assessment of a massively parallel sequencing approach for the
identification of individuals from mass graves of the Spanish Civil War
(1936–1939), Electrophoresis 37 (21) (2016) 2841–2847.
[107] I. Zupanič Pajnič, B. Gornjak Pogorelc, J. Balažic, T. Zupanc, B. Štefanič, Highly
efficient nuclear DNA typing of the World War II skeletal remains using three new
autosomal short tandem repeat amplification kits with the extended European
Standard Set of loci, Croatian Med. J. 53 (1) (2012) 17–23.
[108] T. Leskovar, I. Pajnič, Ž. Geršak, I. Jerman, M. Črešnar, ATR-FTIR spectroscopy
combined with data manipulation as a pre-screening method to assess DNA
preservation in skeletal remains, Forensic Sci. Int. Genet. 44 (2020), 102196.
37
Chapter 3: DNA Extraction from

Incinerated Bone

38
3a. Validation of porcine primers

3a.i Introduction

Human DNA extraction and amplification assays are well developed and readily available, however

accessing human tissue for forensic research can be difficult. By comparison, animal tissue is easy to

source but there is significantly less published research and few commercially available DNA kits.

Before DNA extraction and amplification can be conducted species-specific primers must be

developed, their thermocycling conditions determined, and their general effectiveness tested. As the

use of porcine DNA is most common in agricultural research and the food industry, pre-existing

primers for domestic pigs are all mitochondrial due to the multi copy nature and increased robustness

of mtDNA. For this reason, prior to the commencement of candidacy, two porcine nuclear primer sets

(Table 3.1) were developed by my primary supervisor (DH). Upon starting my candidacy, I tested

these primers, created standards and developed standard curves to allow me to accurately quantify

DNA extracted from porcine bone. I then used these primers to test DNA extraction methods as

documented in the study presented in part b of this chapter.

Table 3.1. Summary of porcine primer sets developed for use in this study.

Sequence (5’-3’) Length Tm GC% Self Product

complementarity Length

Primer pair 1 (long)

Forward CTCTGACCTGAGTCTCCTTT 20 55.89 50.00 5 74
Reverse CAAACACGAGAAAGACTCCA 20 55.64 45.00 3

Primer pair 2 (short)

Forward CTCTGACCTGAGTCTCCTTT 20 55.89 50.00 5 150
Reverse CGGCTTTGTCACACGAG 17 55.94 58.82 5
Details: Sus scrofa isolate TJ Tabasco breed Duroc chromosome 3, Sscrofa 11.1,
products associated with cytoplasmic actin 1

39
3a.ii Methods

Quality control testing of the developed primers began with extraction and standard PCR

amplification of DNA from fresh porcine tissue. A porcine long bone still encased in flesh was

obtained from a local butcher. The surface dermal layer was removed, and a small section of tissue

was collected using sterile forceps and a scalpel. Unlike later extractions involving bone tissue, no

decalcification step was required. A standard Qiagen DNeasy Blood and Tissue kit was used to

extract DNA with the following protocol:

1. Add 40 µl of proteinase K and 160 µl of Buffer ATL to 2 ml tube containing sample

2. Incubate tube at 55 °C overnight using a thermal shaker

3. Centrifuge at 10,000 rpm for 5 min to pellet any remaining undigested material

4. Remove 1 ml of supernatant to new 5 ml tube, add 1000 µl AL buffer

5. Vortex for 15 s and add 1000 µl Ethanol (100%)

6. Vortex for 15 s (critical to avoid formation of precipitate)

7. Pipette 650 µl of solution into DNeasy spin column in 2 ml collection tube

8. Centrifuge at 8000 rpm for 1 min, discard flow-through (hazardous) and reuse collection tube

9. Repeat steps 7-8 until entire sample spun through

10. Place DNeasy spin column in new 2 ml collection tube, add 500 µl Buffer AW1 and

centrifuge for 1 min at 8000 rpm. Discard flow-through and collection tube.

11. Place DNeasy spin column in new 2 ml collection tube, add 500 µl Buffer AW2 and

centrifuge for 3 min at 14,000 rpm to dry DNeasy membrane. Discard flow-through and

collection tube.

12. Place DNeasy mini-spin column in clean 1.5 ml microcentrifuge tube and pipette 100 µl

Buffer AE directly onto DNeasy membrane.

13. Incubate at room temperature for 10 min, then centrifuge for 1 min at 8000 rpm to elute.

14. Transfer DNA extract to clean tube, make a 30 µl working aliquot.

Extractions were performed in a dedicated low-copy laboratory and tools and equipment were

regularly sterilised in a UV cabinet.

40
A standard PCR was performed on the fresh DNA extracts using an Eppendorf® thermal cycler;

contents of the PCR master mix and thermocycling conditions are presented in table 3.2. Two master

mixes were created, one using the short amplicon reverse primer and another using the long amplicon

reverse primer.

Table 3.2. Summary of Plat Taq HiFi Master Mix components and PCR thermocycling conditions.

Plat Taq HiFi Master Mix Thermocycling Conditions

Setup: Volume (µl): Step Duration/temp
ddH2O 14.275 Denaturation: 94 °C hold 2 min
HiFi buffer 10 x 2.5 x30 Cycles:
RSA (10 mg/ml) 2.5 1. Denaturation 94 °C hold 15 s
MgSO4 (50 mM) 1.0 2. Annealing 56 °C hold 15 s
Primer L 1.0 3. Extension 68 °C hold 45 s
Primer H 1.0 Extension: 68 °C hold 10 min
dNTPs (10 mM each) 0.625
Plat Taq HiFi (5 U/µl) 0.1
Extract 2.01
TOTAL VOLUME (per sample): 25.0

A high sensitivity DNA assay was then used to test for specificity of the amplified product and

presence of primer dimer. This was achieved using an Agilent 2100 Bioanalyzer (2021 Agilent

Technologies, Inc.), which uses capillary electrophoresis in a chip-based format allowing for largely

automated analysis of DNA. The Bioanalyzer is ideal for high resolution analysis of small sample

volumes, although the limited sample throughput (12 samples per chip/cartilage) and high cost can be

a major drawback. This technology was ideal for my pilot study as only two primer products (four

replicates for each) were investigated for downstream applications. The resulting electrophoresis

assay was well resolved, with bands matching in each replicate and no evidence of non-target

amplification (Fig. 3.1). Similarly, the electropherogram showed fragments were of the expected

length and non-target peaks were negligible (Fig. 3.2).

41
Fig 3.1. High sensitivity DNA electrophoresis assay. In addition to sample replicates (n=4 per primer

set), negative controls (NC) and extraction blanks (EB) were included.

Fig 3.2. High sensitivity DNA electropherograms for both long (primer pair 1) and short (primer pair

2) amplicons.

42
Post-PCR product was then purified using AMPure paramagnetic beads (Argencourt Bioscience

Corporation), which selectively bind to post-PCR DNA fragments, allowing for non-target material

such as nucleotides, enzymes and excess primers to be removed using simple washing. A Qubit 2.0

fluorometer was subsequently used to determine DNA concentration of the stock solution. Melt-curve

analysis was conducted on the purified product using a Rotor-Gene 3000TM real time thermal cycler

(Corbett Research) to determine optimal thermocycling conditions. Melt curves for each primer pair

were well resolved, and gradient qPCR between 50-60 °C showed 56 °C was the optimal

amplification temperature (Fig 3.3). Peaks corresponded to 80.3 °C for amplicon 2 (primer pair 2) and

83.0 °C for amplicon 1 (primer pair 1).

Step Duration/temp
Denaturation: 94 °C hold 2 min
x35 Cycles:
1. Denaturation 95 °C hold 30 s
2. Annealing 56 °C hold 30 s
3. Extension 68 °C hold 45 s
Extension: 60 °C hold 10 min
MCA: 66-96 °C, 3 °C/s hold 5 s

Fig 3.3. Optimal thermocycling conditions with gradient melt curve analysis step. Subsequent figure

shows cumulative melt peaks for both amplicons.

43
After establishing concentration and thermocycling conditions, a standard curve was created using

purified extract, which was diluted in series (Fig. 3.4). The standard curve was then used to compare

amplification profiles of unknown samples to those of known concentration, along with extraction

blanks and negative controls.

Fig 3.4. Line of best fit representing logarithmic decrease in DNA concentrations at increasing

dilution. Each standard dilution was run in triplicate however, some data were excluded from the

standard curve as not all dilutions were included in the final run.

3a.iii Concluding statement

From this analysis the developed primers were deemed fit for purpose and implemented in the next

study comparing demineralisation procedures for DNA extraction.

44
 3b: Comparison of Bone

Demineralisation Procedures for DNA

Recovery from Burned Remains

Published: Mckinnon, M, & Higgins, D 2020, ‘Comparison of bone demineralisation

procedures for DNA recovery from burned remains’, Forensic Science International: Genetics, vol.

51, doi: https://doi.org/10.1016/j.fsigen.2020.102448

45
 Statement of Authorship
Title of Paper: Comparison of Bone demineralisation procedures for DNA recovery from burned remains

Publication Status: ☒ Published

☐ Accepted for Publication

☐ Submitted for Publication

☐ Unpublished & Unsubmitted work written in Manuscript Style

Publication Details: Published in Forensic Science International: Genetics in March 2021.

Principal Author (Candidate): Meghan Mckinnon

Contribution to the paper: Conceptualization, methodology, formal analysis, investigation,
writing – original draft, editing.
Overall percentage (%): 90%
Certification: This paper reports on original research I conducted during the period of my Higher
Degree by Research candidature and is not subject to any obligations or contractual
agreements with a third party that would constrain its inclusion in this thesis. I am
the primary author of this paper.

Signature & Date: __ _________________________________________

Co-Author Contributions
By signing the Statement of Authorship, each author certifies that:
i. the candidate’s stated contribution to the publication is accurate (as detailed above);
ii. permission is granted for the candidate in include the publication in the thesis; and
iii. the sum of all co-author contributions is equal to 100% less the candidate’s stated contribution.

Co-Author: Denice Higgins (secondary author)

Contribution to the paper: Conceptualization, methodology, investigation, resources, writing –
review & editing, supervision, project administration, funding acquisition.

Signature & Date: __ ________________________________

46
 Forensic Science International: Genetics 51 (2021) 102448

Contents lists available at ScienceDirect

Forensic Science International: Genetics

journal homepage: www.elsevier.com/locate/fsigen

Research paper

Comparison of bone demineralisation procedures for DNA recovery from

burned remains
Meghan Mckinnon a, *, Denice Higgins b
a
Discipline of Anatomy and Pathology, Adelaide Medical School, The University of Adelaide, Adelaide, Australia
b
Forensic Odontology Unit, Adelaide Dental School, The University of Adelaide, Adelaide, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Recovering DNA from modern incinerated bones can be challenging and may require alteration of routine DNA
Burned bone extraction protocols. It has been postulated that incinerated bones share some similarities with ancient bones,
Diagenetic bone including fragmented DNA, surface contamination and highly mineralised structure, all of which can inhibit the
Bone demineralization
successful recovery of genetic material. For this reason, ancient DNA extraction protocols are often used for
DNA extraction protocol
incinerated modern samples; however, their effectiveness is still somewhat unclear. Much of this uncertainty
Forensic DNA recovery
Degraded DNA exists around the demineralisation step of extraction, specifically the length of incubation and retention or
removal of supernatant. As obtaining human samples for forensic research can be challenging, porcine models
(Sus scrofa domesticus) are often used as substitutes. This study developed real time PCR assays for porcine nu­
clear DNA in order to investigate the effects of modified demineralisation protocols on DNA yield from femurs
exposed to either short (60 min) or prolonged (120 min) burning. Gradient PCR results indicated 56 ◦ C was the
ideal amplification temperature for targeted amplicons, with melt curve analysis showing short and long
amplicons corresponded to 80.3 ◦ C and 83 ◦ C peaks respectively. Results of altered extraction protocol showed a
trend towards higher DNA yields from longer demineralisation periods however this was not significant. By
comparison, retaining supernatant post-demineralisation resulted in significantly greater DNA yields compared
to discarding it (P < 0.009). Although DNA content yield decreased with burn duration, the demineralisation
treatment variations appeared to have the same effect for all burn lengths. These results suggest that for
incinerated modern bone retaining the supernatant following demineralisation can dramatically increase DNA
yield.

1. Introduction
Post-mortem human remains may be subjected to a wide variety of
physical and environmental influences resulting in differential preser­
vation, particularly in large-scale disasters [1]. These influences result in
fragmentation, burning, extreme weathering, contamination and com­
ingling of remains [2], with variable impact on DNA content. For victims
of forest, structural and vehicular fires sampling of hard tissue may be
the only way of obtaining viable genetic material for identification, as
soft tissues are usually compromised or unavailable [3,4]. As DNA in
bone is intimately associated with the mineral structure demineralisa­
tion has been shown to be a vital step in liberating DNA from highly
degraded samples [5–8]. Demineralisation involves decreasing the ac­
tivity of inorganic bone calcium (Ca2+), breaking down the mineral
structure to liberate DNA trapped within the crystal bone matrix [9]. As
calcium can inhibit polymerase chain reaction (PCR) its removal will
also enhance downstream processes [10]. Choi et al. [11] compared the
outcome of demineralisation using various acidic and chelating agents
for DNA recovery, concluding that EDTA best preserved genetic mate­
rial. This finding has been supported by many similar studies [12–15].
For ancient bone samples, total demineralisation utilising prolonged
incubation periods (>12 h) has been shown to be the most effective
technique for maximising DNA yields [5–8]. However, this may not be
ideal for all samples, and demineralisation procedures may require
modification depending on sample preservation. Rohland & Hofreighter
[16] suggest that prolonging demineralisation for ancient samples does
not always increase DNA typing success, but state that it can decrease
the amount of undigested bone which may improve DNA yield.

* Corresponding author at: Discipline of Anatomy and Pathology, Adelaide Medical School, The University of Adelaide, Helen Mayo North, Frome Road, Adelaide,
SA, 5000, Australia.
E-mail address: meghan.mckinnon@adelaide.edu.au (M. Mckinnon).

https://doi.org/10.1016/j.fsigen.2020.102448
Received 7 August 2020; Received in revised form 9 November 2020; Accepted 1 December 2020
Available online 26 December 2020
1872-4973/© 2020 Elsevier B.V. All rights reserved.

47
M. Mckinnon and D. Higgins
Prolonged demineralisation has been shown to be critical for ancient
tissues that are highly mineralised, and contain virtually no DNA outside
the protective hydroxyapatite crystalline lattice [17]. By comparison,
modern samples may not require demineralisation as there is usually
sufficient cellular material to provide DNA that is not bound to hy­
droxyapatite. For this reason, modern hard tissue DNA protocols (e.g.
QIAamp DNA Mini Kit by Qiagen) do not include a demineralisation
step. Modern burned bone, however, is believed to present many of the
same structural and compositional changes as diagenetically altered
bone, therefore DNA contained within this tissue may behave in a
similar manner to ancient samples [18]. Not only is there debate over
demineralisation duration, but whether or not the supernatant should be
retained has also been questioned [19]. For ancient samples the super­
natant is often discarded to ensure removal of inhibitory material and
exogenous DNA contamination that may be present [20,21]. However,
for modern bone samples where DNA is also contained within cellular
components, a large amount of target endogenous DNA may also be
discarded with the supernatant [8]. Due to the current lack of estab­
lished protocols for modern burned remains, sampling and deminerali­
sation is often performed ad-hoc, which can be particularly problematic
in high-throughput situations where samples are limited in size and
number and must be processed quickly (i.e. mass disasters).
Due to ethical restrictions on the use of human samples, animal
models are often used as an alternative. Domestic pigs (Sus scrofa
domesticus) are a common substitute for humans due to the similarity in
body size and bone composition [22–24]. Furthermore, the genome of
pigs is relatively similar to humans, which makes them viable substitutes
for DNA based research [22]. In this study we developed PCR assays
targeting porcine specific DNA and investigated demineralisation
duration and supernatant retention or disposal on nuclear DNA (nDNA)
yield from porcine bones exposed to temperatures >100 ◦ C. We
compared DNA yield from samples demineralised using a long incuba­
tion period (12 h), common in ancient DNA protocols, to a shorter (1 h)
period. We also examined a third treatment group where, in addition to
a shorter demineralisation period, the supernatant was retained.
Fig. 1. Flowchart depicting the allocation of samples into treatment groups. Figure created with BioRender.com.
Forensic Science International: Genetics 51 (2021) 102448
2. Methods
2.1. Sample preparation
Three porcine long bones (encased in flesh) were exposed to an open
fire, for either a short (60 min) or long (120 min) duration. No additional
accelerants were used besides naturally occurring bush material. Bone
powder was then obtained:
• Soft tissues removed using scalpels and forceps.
• Surface cleaned of any remaining tissue by alternate gentle rubbing
with Isopropanol (IsoWipes) and 3% sodium hypochlorite solution.
• Bone sampled using a hand-held drill (pre-cleaned with 3 % sodium
hypochlorite) fitted with a new 3.5 mm drill bit for each sample.
Shavings collected on aluminium foil in a clean petri dish.
• Drill speed <100 RPM to reduce heat generation that could denature
the DNA [25].
• 180 mg of bone shavings was obtained from each compact bone then
divided into 9 × 2 ml tubes (Fig. 1).
• Samples stored at − 20 ◦ C immediately after drilling.
2.2. Silica column extraction
Extractions were performed in a dedicated low-copy pre-PCR lab in a
fume hood. All surfaces and equipment pre cleaned with sodium hypo­
chlorite. Extractions performed in triplicate (as was amplification and
quantification). A negative control was included in each group.
For creation of qPCR standards DNA was extracted from fresh
porcine soft tissue using QIAamp DNA Mini Kit following the manu­
facturer’s instructions (Qiagen).
For bone samples, the same extraction process was used with the
addition of a demineralisation step prior to digestion following Loreille
et al. [8] with slight modification as described below:
• In a 2 ml tube 950 μl of 0.5 M EDTA was added to 20 mg of ground
bone sample.
48
M. Mckinnon and D. Higgins Forensic Science International: Genetics 51 (2021) 102448

• Samples placed on a thermal shaker, incubated at 55 ◦ C and using a Qubit 2.0 fluorometer (InvitrogenTM), then diluted in series into
continually mixed for; six 45 μl standard aliquots (Table 2).
- 1 h (Qiagen standard incubation, treatment groups 1 & 2), or
- 12 h overnight (prolonged incubation treatment group 3). 2.5. Quantification of samples
• EDTA supernatant was subsequently either:
- Retained (treatment group 1), or Sample extracts were run alongside both negative and positive (PCR
- Discarded (treatment groups 2 & 3) standards) controls using both primer pair assays (short reverse primer
• Samples were digested using EDTA (950 μl), Proteinase K (40 μl), and and long reverse primer). For each assay 27 samples of unknown con­
Buffer ATL (160 μl). centration were analysed, resulting in a total of 54 samples for each of
• cRNA was added with Buffer AL (3 μl cRNA per 1000 μl Buffer AL). the three burn groups (Fig. 1). Thermocycling was conducted using the
• Samples were filtered through spin columns in multiple fractions of optimised protocol described earlier, and Rotor-Gene software was used
650 μl. to apply the standard curve to the unknown samples. Absolute quanti­
fication values were collated and analysed in IMB SPSS statistics [27].
2.3. Primer design and efficacy
3. Results & discussion
A single forward and two reverse porcine specific primers were
The premise of this study centres around the comparability of archaic
designed to target two amplicons (one short and one long) in a non-
bone (defined here as deposited >200 years) to thermally altered bone,
repeating area of the porcine nuclear genome (Table 1). Two ampli­
the assumption being that in both cases exogenous DNA is released
cons were selected for the purpose of quality control. Along with
during demineralisation whilst minuscule quantities of endogenous
effectively doubling the sample size, using two amplicons of different
material will be bound to tight crystal aggregates of HAp [28]. Ancient
lengths mean that in cases of severe degradation, shorter fragments may
DNA extraction techniques have been designed with this in mind and are
still be detectable [26]. PCR primers were checked for specificity using
optimised to remove external contaminants (by discarding initial su­
the NCBI basic local alignment search tool (https://www-ncbi-nlm-nih
pernatant) whilst liberating bound material (by prolonged deminerali­
-gov.proxy.library.adelaide.edu.au/tools/primer-blast/).
sation) [8,9]. Similar to ancient material, we expected a great deal of
To avoid complications of multiplexing primers independent assays
external contamination to be present in thermally altered bone. We
were developed for short and long amplicons. The optimal annealing
predicted that like archaic bones, prolonged demineralisation would be
temperature for each PCR assay was determined using gradient PCR. All
required to liberate DNA, as bone becomes increasingly crystalline when
reactions were performed with the Rotor-Gene 3000™ real time thermal
heated making DNA less accessible [17,29]. The reality behind this is
cycler (Corbett Research). PCR cycling conditions were: initial dena­
unclear, and previous studies have produced mixed results [19,30].
turation at 94 ◦ C for 2 min, followed by 35 cycles of denaturation at
Although demineralisation is vital for dense, degraded and contami­
95 ◦ C for 30 s, annealing at 56 ◦ C for 30 s, primer extension at 68 ◦ C for
nated samples, there is some uncertainty on how this process should be
45 s, with a final extension step at 60 ◦ C for 10 min, followed by Melt
conducted. The primary objective of this study was to optimize demin­
Curve Analysis [MCA] (66 ◦ C–96 ◦ C, 3 ◦ C/s). The specificity of primers
eralisation techniques for maximum yield of nuclear DNA from ther­
to a single binding site was assessed using the post qPCR melt curve to
mally altered porcine bone – specifically EDTA incubation period length
visualize the dissociation kinetics. The assays were conducted in a 15 μl
and impact of retaining demineralised supernatant.
total reaction volume containing 0.5 μl of both the forward and reverse
primer (10 mM), 0.5 μl Rabbit Serum Albumin [RSA](10 mg/mL) and
3.1. Development of qPCR assays
1 μl porcine genomic DNA extract. PowerUp™ SYBR™ Green Master
Mix (Invitrogen™) made up the final 7.5 μl of the total reaction volume.
Gradient PCR identified 56 ◦ C as best temperature for targeted
amplicons, amplification at this temperature resulted in highly specific,
2.4. Creation of qPCR standards identical peaks for all triplicates. The short and the long amplicons
corresponded to peaks at 80.3 ◦ C and 83 ◦ C, respectively, which were
Standards for each primer pair were created by amplifying DNA well resolved in the Melt Curve Analysis profile. The standard curves
extracted from porcine soft tissue using an Eppendorf® thermocycler. A comprised a six-point, dilution series for each primer set, with each
reaction volume of 25 μl was used, consisting of 14.2 μl H2O, 2.5 μl HiFi dilution level performed in three replicates. Reliable quantification was
buffer, 2.5 μl RSA (10 mg/mL), 1.0 μl MgSO4 (50 mM), 1.0 μl forward/ established for the range of input DNA shown in Table 2 per reaction,
reverse primers (10μM), 0.6 μl dNTPs (10 mM), 0.1 μl Platinum® Taq with an acceptable linear range (R2 = 0.0999). For both DNA targets the
HiFi (5U/μl) (Invitrogen™) and 2.0 μl extract. Post-PCR product was amplification was repeatable and reproducible over the three replicates.
diluted to 1:10 concentration and run on a bioanalyser (Agilent High
Sensitivity DNA kit, 2100 expert software). The high sensitivity DNA
assay showed amplified sequences were identical in length to target
amplicons. Most samples tested were free of primer dimer, and the few
that were not, contained only negligible amounts which were removed Table 2
using AMPure paramagnetic beads (Argencourt Bioscience Corporation, DNA fragment copies within neat standard (assessed using Qubit 2.0 fluo­
Beverly, Massachusetts). Neat control concentrations were quantified rometery) and copies in subsequent serial dilutions.
Serial Short Short Copies/μl Long Long Copies/μl
Table 1 Dilution of fragment fragment
Standards conc. (ng/ conc. (ng/
Primer sequences used for amplification of target DNA in qPCR.
μl) μl)
Primer Name: Sequence (5’-3’) Primer Amplicon
Standard 6 2.4 60,100,000,000 5.6 69,200,000,000
length Length
(Neat)
Forward CTCTGACCTGAGTCTCCTTT 20 Standard 5 0.24 6010,000,000 0.56 6920,000,000
Reverse short CGGCTTTGTCACACGAG 17 74 Standard 4 0.024 601,000,000 0.056 692,000,000
(fragment 1) Standard 3 0.0024 60,100,000 0.0056 69,200,000
Reverse long CAAACACGAGAAAGACTCCA 20 150 Standard 2 0.00024 6,010,000 0.00056 6,920,000
(fragment 2) Standard 1 0.000024 601,000 0.000056 692,000

49
M. Mckinnon and D. Higgins Forensic Science International: Genetics 51 (2021) 102448

3.2. Effect of DNA extraction protocol Table 3

Mann-Whitney test multiple comparisons showing specific relationships be­
In all burn states retaining the supernatant and demineralising for 1 h tween each EDTA treatment within each burn group. Also included is the
was the most effective technique, resulting in the highest number of average percentage difference in DNA yield from one treatment to the other.
fragment copies recovered by both assays (Figs. 2). Discarding the su­ 12 h + discard vs. 12 h + discard vs. 1 h + discard vs.
pernatant had the greatest impact on results, and substantially reduced 1 h + discard 1 h + retained 1 h + retained
the quantity of DNA recovered in both burned and unburned samples. Fragment Length: Short Long Short Long Short Long
Comparatively, demineralisation duration had little effect on the Unburned 0.730 0.077 0.000* 0.000* 0.094 0.005*
quantity of DNA recovered except in unburned samples, where a longer 60 min. burn 0.796 0.546 0.000* 0.321 0.000* 0.743
demineralisation resulted in a slightly improved DNA yield. 120 min. burn 0.387 0.190 0.000* 0.000* 0.000* 0.000*
Our data was not normally distributed; hence a non-parametric Grouped 0.727 0.701 0.000* 0.000* 0.000* 0.009*
Concentration ±16 % ±45 % ±91 % ±127 % ±64 % ±56 %
approach was used. Analysis of variance shows significant differences
diff
(P < 0.05) in DNA yield between EDTA treatments in all burn groups
*
with the exception of long fragments in the 60-minute burn group Significant (P > 0.05).
(χ2 = 1.055, P = 0.59). Post-hoc Mann-Whitney tests (Table 3) showed
that within all burn groups there was no significant effect of incubation 3.3. Effect of burn duration
time on DNA yield when supernatant was discarded. In nearly all cases
retaining the supernatant significantly increased DNA copies. Although In this study although DNA yield decreased as expected with longer
not significant, a general increase of short (+16 %) and long (+45 %) durations of burning this was not always significant (Table 4). In all
fragment DNA was apparent where incubation was increased from 1 to EDTA treatments unburned samples contained significantly more long
12 h (supernatant discarded). Comparatively, incubating for 1 h and and short DNA fragment copies than burned samples (p < 0.01).
retaining the supernatant significantly increased DNA yield when Comparatively, DNA yield from 60 min and 120 min were not signifi­
compared to the prolonged 12 h incubation and discard of supernatant cantly different (p > 0.29) with the exception of EDTA treatment 1 (1 h
(short = +91 %, long = +127 %) and when compared to 1 h incubation decalcification, supernatant retained) where short fragment yield was
and discard of supernatant (short = +64 %, long = +56 %). significantly greater in the 60 min burn group than the 120 min.
Overall, increased demineralisation duration (12 h) showed no DNA has a strong affinity for inorganic bone due to ionic binding
improvement over the shorter duration (1 h). Whilst a longer deminer­ between positively charged calcium cations (Ca2+) in hydroxyapatite
alisation period appeared to perform better in unburned samples, this and negatively charged phosphate groups in the DNA backbone [32,33].
was not significant. In regard to disposal of demineralised supernatant, As bone crystallisation changes dramatically when exposed to high
our results clearly show a significant improvement in DNA fragment and/or prolonged temperatures, it is logical to assume this process
yield from samples where supernatant is retained. Prado et al., 2002 would affect bound DNA. Studies show that when exposed to tempera­
explains that cellular lysis EDTA treatment causes defragmentation of tures of 500− 600 ◦ C bone enters a highly crystalline phase, with hy­
the bone matrix and subsequent release of DNA to extracellular medium, droxyapatite crystal size increasing dramatically [18,34–36]. It is likely
which is then lost during subsequent washes. Prado et al. found that co-
extracting supernatant with the undigested pellet ensured retention of Table 4
usually discarded DNA. This aligns with the current study, where in Mann-Whitney test multiple comparisons showing specific relationships be­
some cases retaining supernatant more than doubled DNA fragment tween each duration of burning within each EDTA treatment group.
copies. In some situations, the bone pellet may be digested in its entirety Unburned vs. Unburned vs. 60 min vs.
(total demineralisation) to yield more DNA [8]. This method ensures all 60 min 120 min 120 min
DNA within a sample is extracted, however with particularly degra­ Fragment Length: Short Long Short Long Short Long
ded/dense samples this can be time consuming, and often results in
EDTA Treatment 1 0.000* 0.006* 0.000* 0.000* 0.387 0.666
co-extraction of surface contamination [5]. In such cases, disposal of
EDTA Treatment 2 0.001* 0.014* 0.001* 0.000* 0.931 0.297
undigested material may result in an unavoidable loss of DNA [31,16]. EDTA Treatment 3 0.004* 0.001* 0.000* 0.000* 0.031* 0.423
*
Significant (P > 0.05).

Fig. 2. Comparison of mean endogenous A) short and B) long nDNA fragment copies between EDTA treatment groups with standard error bars.

50
M. Mckinnon and D. Higgins
that the samples in the current study did not reach the 600 ◦ C point of
recrystallization, meaning that despite incineration of the outer soft
tissue the interior was still relatively well protected even at the longer
duration. This is where limitations of the current study must be
considered. The very nature of attempting to recreate a realistic model of
burning meant that many variables could not be controlled. Although
consistency was maintained between samples (i.e. same accelerants
used, same weather conditions) the exact temperature that samples were
exposed to could not be measured. Whole porcine hocks were used to
closely mimic the general dimensions of human bones and tissue
thickness, the logistics of which meant an electric muffle furnace could
not be used – additionally using a homogeneous heat source such as a
furnace would not accurately represent a disaster event. Although less
than ideal, the duration of burning presented an opportunity to assess
diversity that may occur amongst burned samples in identification cases.
Based on our results, we can make some general recommendations
regarding demineralisation methodology for samples where bone has
not yet reached calcining temperatures (>600 ◦ C). During deminerali­
sation EDTA supernatant should be retained to prevent accidental
disposal of genetic material. Where total demineralisation is not
possible, undigested bone pellets should be retained for later analysis. If
standard demineralisation protocol does not produce acceptable results,
the undigested pellet can be processed as a last resort. Although this
study strongly supports retention of supernatant during demineralisa­
tion, there are limitations that must be counteracted. Fore mostly the
retention of surface contaminants, which is the main reason for disposal
of supernatant for archaic samples [20]. The use of albumin such as BSA
(in our case RSA) can significantly reduce inhibitor interference of
polymerases and is commonly used for archaic samples [37]. Decreased
inhibition also removes the need for excessive post-PCR purification,
which can be time consuming [38]. In addition to RSA, we opted to
include carrier RNA (cRNA). Although not directly linked to inhibition
and contamination of samples, using carrier molecules in silica-based
extraction can significantly increase yield where target DNA is ex­
pected to exist in low quantities (e.g. archaic/degraded material). The
use of cRNA is still somewhat debatable, with some studies showing no
improvement in DNA yield with its addition [39]. However, it has also
been suggested that other constituents (such as EDTA) could affect the
interaction of cRNA with silica columns, decreasing its effectiveness
[40]. These processes are not necessary for all sample types, however in
cases where samples are highly degraded those with the time and re­
sources may benefit greatly.
4. Conclusion
This study developed a real time PCR assay for porcine nuclear DNA
and demonstrated that modern burnt porcine bone did not respond to
extraction techniques in the manner of archaic bone, but more appeared
to mimic fresh bone. An increase in the duration of demineralisation was
shown to have little effect on the yield of DNA recovered. Under the
specific conditions investigated, the efficiency of DNA recovery from
burned remains was improved by co-extraction of the demineralised
supernatant and a brief (1 h) EDTA incubation period. Although similar
changes in bone crystallinity may occur in both burned and archaic
samples, the interaction between mineral and DNA may vary between
these sample types. Hence, these sample types will need differing
extraction techniques to optimise DNA yield. In future work, prolonging
burn duration to reach calcining temperatures, along with the use of
accurate temperature sensors could significantly improve our
understanding.
Funding
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
Forensic Science International: Genetics 51 (2021) 102448
Availability of data and material
The raw dataset generated for the purpose of this study is available
on reasonable request via the corresponding author.
CRediT authorship contribution statement
Meghan Mckinnon: Conceptualization, Methodology, Formal
analysis, Investigation, Data curation, Writing - original draft, Writing -
review & editing. Denice Higgins: Conceptualization, Methodology,
Investigation, Resources, Writing - review & editing, Supervision, Proj­
ect administration, Funding acquisition.
Declaration of Competing Interest
The authors report no declarations of interest.
Acknowledgements
The authors would like to thank the Advanced DNA, Identification
and Forensic Facility (ADIFF) at the University of Adelaide for access to
DNA extraction and PCR setup laboratories, and Dr Jennifer Young for
her technical assistance.
References
[1] B. Budowle, F. Bieber, A. Eisenberg, Forensic aspects of mass disasters: strategic
considerations for DNA-based human identification, Leg. Med. 7 (4) (2005)
230–243.
[2] A. Mundorff, R. Shaler, E. Bieschke, E. Mar-Cash, Marrying anthropology and DNA:
essential for solving complex commingling problems in cases of extreme
fragmentation, in: B. Adams, J. Byrd (Eds.), Commingled Human Remains,
Academic Press, San Diego, 2014, pp. 257–273.
[3] A. Alonso, P. Martin, C. Albarran, et al., Challenges of DNA profiling in mass
disaster investigations, Croat. Med. J. 46 (4) (2005) 540–548.
[4] C. Lassen, S. Hummel, B. Herrmann, Comparison of DNA extraction and
amplification from ancient human bone and mummified soft tissue, Int. J. Legal
Med. 107 (3) (1994) 152–155.
[5] J. Jakubowska, A. Maciejewska, R. Pawlowski, Comparison of three methods of
DNA extraction from human bones with different degrees of degradation, Int. J.
Legal Med. 126 (2012) 173–178.
[6] D. Vanek, M. Silerova, V. Urbanova, et al., Genomic DNA extraction protocols for
bone samples: the comparison of Qiagen and Zymo Research spin columns,
Forensic Sci. Int. Genet. Suppl. Ser. 3 (1) (2011) e397–e398.
[7] S. Seo, A. Zhang, H. Kim, J. Yi, H. Lee, D. Shin, S. Lee, Technical note: Efficiency of
total demineralization and ion-exchange column for DNA extraction from bone,
Am. J. Phys. Anthropol. 141 (2010) 158–162.
[8] O. Loreille, T. Diegoli, J. Irwin, et al., High efficiency DNA extraction from bone by
total demineralization, Forensic Sci. Int. Genet. 1 (2007) 191–195.
[9] D. Higgins, A.B. Rohrlach, J. Kaidonis, et al., Differential nuclear and
mitochondrial DNA preservation in post-mortem teeth with implications for
forensic and ancient DNA studies, PLoS One 10 (5) (2015) 1–17.
[10] K. Opel, D. Chung, B. McCord, A study of PCR inhibition mechanisms using real
time PCR, J. Forensic Sci. 55 (1) (2010) 25–33.
[11] S. Choi, S. Hong, S. Yoon, Proposal of an appropriate demineralisation method of
bone marrow biopsy specimens in the era of expanding genetic molecular study,
J. Pathol. Transl. Med. 49 (3) (2015) 236–242.
[12] V. Singh, R. Salunga, V. Huang, et al., Analysis of the effect of various
demineralisation agents on the quantity and quality of nucleic acid (DNA and RNA)
recovered from bone biopsies, Ann. Diagn. Pathol. 17 (4) (2013) 322–326.
[13] R. Brown, J. Edwards, A. Bartlett, et al., Routine acid demineralisation of bone
marrow samples can preserve DNA for FISH and CGH studies in metastatic prostate
cancer, J. Histochem. Cytochem. 50 (1) (2002) 113–115.
[14] S. Tinling, R. Giberson, R. Kullar, Microwave exposure increases bone
demineralization rate independent of temperature, J. Microsc. 215 (3) (2004)
230–235.
[15] J. Alers, P. Krijtenburg, K. Vissers, H. Van Dekken, Effect of bone demineralisation
procedures on DNA in situ hybridization and comparative genomic hybridization:
EDTA is highly preferable to a routinely used acid decalcifier, J. Histochem.
Cytochem. 47 (5) (1999) 703–709.
[16] N. Rohland, M. Hofreiter, Comparison and optimization of ancient DNA extraction,
Biotechniques 42 (3) (2007) 343–352.
[17] C. Mulligan, Isolation and analysis of DNA from archaeological, clinical, and
natural history specimens, in: E. Zimmer, E. Roalson (Eds.), Methods in
Enzymology, vol. 395, Academic Press, 2005, pp. 87–103.
[18] P. Shipman, G. Foster, M. Schoeninger, Burnt bones and teeth: an experimental
study of color, morphology, crystal structure and shrinkage, J. Archaeol. Sci. 11 (4)
(1984) 307–325.
51
M. Mckinnon and D. Higgins
[19] M. Fondevila, C. Phillips, N. Naveran, et al., Case report: identification of skeletal
remains using short-amplicon marker analysis of severely degraded DNA extracted
from a decomposed and charred femur, Forensic Sci. Int. Genet. 2 (3) (2008)
212–218.
[20] A. Mohammadi, A. Ghorbani, M. Khafaei, et al., A new and efficient method for
DNA extraction from human skeletal remains usable in DNA typing, J. Appl.
Biotechnol. Rep. 4 (2017) 609–614.
[21] P. Damgaard, A. Margaryan, H. Schroeder, et al., Improving access to endogenous
DNA in ancient bones and teeth, Sci. Rep. 5 (2015), pp. 11184.
[22] J. Fadista, M. Nygaard, E. Holm, et al., A snapshot of CNVs in the pig genome, PLoS
One 3 (2008) pp. e3916.
[23] E. Hart, M. Caccamo, J. Harrow, et al., Lessons learned from the initial sequencing
of the pig genome: comparative analysis of an 8 Mb region of pig chromosome 17,
Genome Biol. 8 (2007).
[24] A. Pearce, R. Richards, S. Milz, et al., Animal models for implant biomaterial
research in bone: a review, Eur. Cell. Mater. 13 (1) (2007) 1–10.
[25] C. Adler, W. Haak, Donlon, et al., Survival and recovery of DNA from ancient teeth
and bones, J. Archaeol. Sci. 38 (2011) 956–964.
[26] M. Colotte, V. Couallier, S. Tuffet, J. Bonnet, Simultaneous assessment of average
fragment size and amount in minute samples of degraded DNA, Anal. Biochem. 388
(2) (2009) 345–347.
[27] IBM Corp. Released, IBM SPSS Statistics for Windows, Version 24.0, IBM Corp.,
Armonk, NY, 2016.
[28] J. Dissing, M. Kristinsdottir, C. Friis, On the elimination of extraneous DNA in fossil
human teeth with hypochlorite, J. Archaeol. Sci. 35 (6) (2008) 1445–1452.
[29] M. Salamon, N. Tuross, B. Arensburg, S. Weiner, Relatively well preserved DNA is
present in the crystal aggregates of fossil bones, Proc. Natl. Acad. Sci. 102 (39)
(2005) 13783–13788.
[30] T. Schwark, A. Heinrich, A. Preuße-Prange, N. von Wurmb-Schwark, Reliable
genetic identification of burnt human remains, Forensic Sci. Int. Genet. 5 (5)
(2011) 393–399.
Forensic Science International: Genetics 51 (2021) 102448
[31] P. Campos, O. Craig, G. Turner-Walker, et al., DNA in ancient bone–where is it
located and how should we extract it? Ann. Anat. Anatomischer Anzeiger 194 (1)
(2012) 7–16.
[32] E. Paget, L. Monrozier, P. Simonet, Adsorption of DNA on clay minerals: protection
against DNaseI and influence on gene transfer, FEMS Microbiol. Lett. 97 (1–2)
(1992) 31–39.
[33] B. Busse, M. Hahn, M. Soltau, et al., Increased calcium content and inhomogeneity
of mineralization render bone toughness in osteoporosis: mineralization,
morphology and biomechanics of human single trabeculae, Bone 45 (6) (2009)
1034–1043.
[34] M. Mckinnon, M. Henneberg, E. Simpson, D. Higgins, A Comparison of crystal
structure in fresh, burned and archaic bone–implications for forensic sampling,
Forensic Sci. Int. (2020), pp. 110328.
[35] S. Etok, E. Valsami-Jones, Wess, et al., Structural and chemical changes of
thermally treated bone apatite, J. Mater. Sci. 42 (23) (2007) 9807–9816.
[36] J. Hiller, T. Thompson, M. Evison, et al., Bone mineral change during experimental
heating: an X-ray scattering investigation, Biomaterials 24 (2003) 5091–5097.
[37] E. Farell, G. Alexandre, Bovine serum albumin further enhances the effects of
organic solvents on increased yield of polymerase chain reaction of GC-rich
templates, BMC Res. Notes 5 (257) (2012).
[38] C. Kreader, Relief of amplification inhibition in PCR with bovine serum albumin or
T4 gene 32 protein, Appl. Environ. Microbiol. 62 (3) (1996) 1102–1106.
[39] P. Reddy, M. Bhavanishankar, J. Bhagavatula, et al., Improved methods of
carnivore faecal sample preservation, DNA extraction and quantification for
accurate genotyping of wild tigers, PLoS One 7 (10) (2012) pp. e46732.
[40] D. Higgins, J. Kaidonis, G. Townsend, et al., Evaluation of carrier RNA and low
volume demineralization for recovery of nuclear DNA from human teeth, Forensic
Sci. Med. Pathol. 10 (2014) 56–61.
52
 Chapter 4: A Comparison of Crystal

Structure in Fresh, Burned and Archaic

Bone – Implications for Forensic

Sampling

Published: Mckinnon, M, Henneberg, M, Simpson, E, Higgins, D 2020, ‘A comparison of crystal

structure in fresh, burned and archaic bone – Implications for forensic sampling’, Forensic Science

International, vol. 313, doi: https://doi.org/10.1016/j.forsciint.2020.110328

53
 Statement of Authorship
Title of Paper: A comparison of crystal structure in fresh, burned and archaic bone – implications for
forensic sampling

Publication Status: ☒ Published

☐ Accepted for Publication

☐ Submitted for Publication

☐ Unpublished & Unsubmitted work written in Manuscript Style

Publication Details: Published in Forensic Science International in August 2020.

Principal Author (Candidate): Meghan Mckinnon

Contribution to the paper: Conceptualization, methodology, formal analysis, investigation, writing
– original draft, writing – editing.
Overall percentage (%): 85%
Certification: This paper reports on original research I conducted during the period of my Higher
Degree by Research candidature and is not subject to any obligations or contractual
agreements with a third party that would constrain its inclusion in this thesis. I am
the primary author of this paper.
Signature & Date: ___________________ 23.06.21
_______________________

Co-Author Contributions
By signing the Statement of Authorship, each author certifies that:
i. the candidate’s stated contribution to the publication is accurate (as detailed above);
ii. permission is granted for the candidate in include the publication in the thesis; and
iii. the sum of all co-author contributions is equal to 100% less the candidate’s stated contribution.
Co-Author: Maciej Henneberg
Contribution to the paper: Conceptualization, methodology, writing – review & editing,
supervision, project administration, funding acquisition.

Signature & Date: __ 23.06.21

________________________________
Co-Author: Ellie Simpson
Contribution to the paper: Conceptualization, methodology, resources, writing – review & editing,
supervision, project administration.

23.06.21
Signature & Date: __ _____________________________________
Co-Author: Denice Higgins (secondary author)
Contribution to the paper: Conceptualization, methodology, investigation, resources, writing –
review & editing, supervision, project administration, funding acquisition.

Signature & Date: ________ 23.06.21

_______________________________

54
 Forensic Science International 313 (2020) 110328

Contents lists available at ScienceDirect

Forensic Science International

journal homepage: www.elsevier.com/locate/forsciint

A comparison of crystal structure in fresh, burned and archaic

bone – Implications for forensic sampling
Meghan Mckinnona,* , Maciej Hennebergb , Ellie Simpsonc, Denice Higginsd
a
Discipline of Anatomy and Pathology, Adelaide Medical School, the University of Adelaide, Adelaide, Australia
b
Discipline of Anatomy and Pathology, Adelaide Medical School, the University of Adelaide, Frome Road, Adelaide, SA 5000, Australia
c
Forensic Science SA, Adelaide, Australia
d
Forensic Odontology Unit, Adelaide Dental School, the University of Adelaide, Adelaide, Australia

A R T I C L E I N F O A B S T R A C T

Article history: Standard protocols for extracting DNA from bone are variable and are largely dependent on the state of
Received 22 March 2020 preservation. In archaic samples, endogenous DNA is believed to be tightly bound to crystal aggregates in
Received in revised form 3 May 2020 the Hydroxyapatite (HAp) matrix requiring prolonged demineralisation to allow its release. By
Accepted 4 May 2020
comparison, fresh bone contains abundant cellular material, discounting the need for demineralisation.
Available online 12 May 2020
Recommendations for incinerated bone, specifically how viable sampling sites should be selected and the
ideal techniques for DNA recovery are unclear, and the protocol used is often selected based on
Keywords:
macroscopic sample appearance.
Bone hydroxyapatite recrystallization
Burned bone
It has been postulated that like archaic bone, burned bone is ‘highly degraded’ and therefore aDNA
Diagenetic bone techniques may present better results for DNA recovery than using fresh protocols. However, little
DNA recovery research has been undertaken comparing the crystal structure of burnt, fresh and archaic bone. This
X-Ray powder diffraction study uses a combination of XRPD and SEM analysis to compare the crystalline profile and microscopic
Scanning electron microscopy appearance of burned bone subjected to temperatures ranging from 100–1000
C, with archaic and fresh
samples. Although macroscopically visually different, fresh samples and samples heated up to 500
C
showed no microscopic differences or significant changes in crystallinity. By comparison, samples heated
above 500
C became significantly more crystalline, with HAp crystal size increasing dramatically.
Archaic samples were different again, more closely resembling the amorphous fresh samples than the
highly crystalline incinerated samples. These results suggests that, potentially, samples burned at 500
C
or lower can be treated as fresh samples, whilst samples exposed to higher temperatures may require
adapted protocols. Whether or not these highly burned samples require demineralisation needs to be
investigated.
Crown Copyright © 2020 Published by Elsevier B.V. All rights reserved.

1. Introduction influence the likelihood of recovering useable genetic information

The successful recovery of DNA from degraded remains is a
challenge faced in various fields of study including palaeobiology,
anthropology and forensic science. In many cases, bones and teeth
are the only surviving materials available for analysis. Hydroxyap-
atite (Ca10(PO4)6(OH)2) is a highly crystalline substance consisting
primarily of calcium and phosphate that is the basis of all forms of
biological apatite [1]. The term “bioapatite” is often used to refer to
the less crystalline, carbonate-containing form of hydroxyapatite
that can be found in living bone. Whilst hydroxyapatite (HAp) is
the main inorganic component of bone, collagen makes up the
primary organic component [2]. The ratio of these components can
* Corresponding author.
E-mail address: meghan.mckinnon@adelaide.edu.au (M. Mckinnon).
from the sample [3–5]. Under different taphonomic conditions the
structure of bone can be radically altered hence the techniques
used to extract DNA may need varying dependant on preservation.
In some instances skeletal remains may have been subjected to
burning or thermal damage which can change the structure of
bone at both the primary (histological) and secondary (macro-
scopic) levels, and severely decrease the availability of organic
material. Under extreme temperatures (>500
C) HAp crystals have
been shown to increase in size, whilst organic content decreases
[6]. It has been suggested that the mechanisms by which HAp
crystals enlarge under higher temperatures may be similar to those
in bones that have undergone prolonged natural diagenesis - for
simplicity’s sake from here on in we will be using the term
“archaic” to refer to bones deposited for at least 200 years that
show evidence of digenetic changes (eg. Mineral loss and
replacement) [7]. Harbeck et al. demonstrated complete loss of

http://dx.doi.org/10.1016/j.forsciint.2020.110328
0379-0738/Crown Copyright © 2020 Published by Elsevier B.V. All rights reserved.

55
 M. Mckinnon et al. / Forensic Science International 313 (2020) 110328
the organic content of bone at approximately 400
C [8], however
this threshold is subjective as in most forensic cases the length of
burning is unknown. Tissue thickness, which is a highly variable,
can delay the degradational effects of burning [9], and some
studies have found protein components still present in bone
burned at 600
C and 700
C [10,11]. Additionally, variables such as
bone density, duration of burning and fuel source could also impact
the exact temperature at which organic material is lost. As the
temperature interval at which organic material is present could
overlap bone recrystallization temperatures (500
C), potentially
DNA could still be found within the HAp matrix at these higher
temperatures.
The assumption that modern degraded bone is similar in
crystalline structure to archaic bone has led to the use of
conservative ancient DNA extraction techniques in some forensic
casework. Decalcification is vital in DNA extraction from archaic
bone, due to the potential binding of DNA to mineral [12–15]. It is
believed that virtually all endogenous DNA in archaic remains is
contained within dense HAp crystal aggregates which must be
disrupted to obtain a usable genetic profile [16]. Decalcification
requires prolonged incubation in an acidic agent (commonly
ethylenediaminetetraacetic acid), increasing the time needed for
the DNA extraction process. Decalcification is generally not
performed on modern/fresh bone as DNA is readily available
within the organic components, and decalcification can actually
result in a loss of vital genetic material [17,18,11]. It is possible that
the destructive processes associated with high temperatures may
alter the crystalline structure of modern bone in a way that
resembles archaic bone indicating that ancient DNA decalcification
protocols would be more appropriate for these samples than fresh
sample protocols. However, if this is not the case a prolonged
demineralisation period may be inappropriate and could actually
reduce the effectiveness of DNA recovery. To investigate this,
the crystalline profiles of modern and ancient bones need to
be compared. If significant changes in bone crystalline structure
are detected following burning, knowing the point at which these
changes occur and the temperature at which the crystalline
structure in burned bone resembles that in fresh bone could
be essential in determining which protocol is appropriate for a
specific sample.
The rate and extent of bone degradation following death is
influenced by environmental factors. For cases of prolonged soil
exposure, the activity of ions within soil can be considered a
driving force behind the destruction or preservation of both
inorganic and organic components. White and Hannus [19] state
that calcium (Ca2+) and phosphorous monoxide (PO) ions exist in a
state of equilibrium with HAp and disturbing this balance can
result in differing degrees of degradation. Microorganisms
decompose collagen in the presence of water and oxygen resulting
in increased carbon dioxide (CO2), bicarbonate (HCO3) and
hydrogen (H+) ions in the soil [20]. In calcium-rich soils the rate
of bone degradation will be decreased as Ca2+ ions can be absorbed
into the bone, thereby replacing ions leeched out into the soil and
stabilising the HAp structure; by this same mechanism, the rate of
degradation of bones deposited in water would be different again
[19]. This has implications for DNA preservation through HAp Ca2+
ionic binding [21,22].
In some ways heat-induced degradation is similar to prolonged
diagenesis, as both result in the initial loss of organic collagen
whilst the inorganic component survives until later stages. Bone
heating eliminates hydroxide (OH-) groups through dehydration,
and in cases where a greater concentration of calcium is present
Calcium Oxide (CaO) can be produced as a secondary by-product at
higher temperatures [23]. Although heat does not radically change
the elemental composition of HAp the interatomic arrangement
can be altered resulting in a change in “phase”. This phase change
can be quantified using X-Ray Powder Diffraction (XRPD) analysis
[24] which examines the scatter patterns of monochromatic X-ray
beams created by the lattice planes of a sample. The process
produces a peak profile unique to the angle of scattering. The
intensities (heights) of the peaks reflect the positioning of atoms
within the crystal lattice [25]. The profiles produced by XRPD can
appear as smooth, wave-like hills, sharp peaks or a combination.
Materials that produce a diffraction pattern of continuous sharp
peaks are crystalline in nature, whereas those with single wavelike
patterns contain amorphous materials [26,27]. It is not uncommon
to see a combination of these peak types, particularly in samples
containing both inorganic and organic material [24]. Once a
diffraction pattern has been produced for the sample of interest,
published reference patterns from specific compounds can be
aligned with the target profile peaks to identify elemental
composition [28]. Once the elements are identified, their crystal
size can be calculated using the X-ray wavelength, peak position
(2u) and the width of a target peak at half its full height [Full Width
at Half Maximum (FWHM)] (Fig. 1).
XRPD analysis can be supplemented by semi-quantitative
analysis of the crystalline composition using techniques such as
Scanning Electron Microscopy (SEM). Unlike XRPD analysis which
provides information on the full volume of a powdered sample,
SEM only allows visual and elemental analysis to be studied at
specific regions. Secondary Electron (SE) images are produced
when the electron beam scans the sample and is reflected off to
produce a topographical image of the samples’ surface. Addition-
ally, Energy-Dispersive X-ray spectroscopy (EDX) can provide
elemental and chemical analysis. In EDX the electron beam hits the
sample causing its atoms to become energised. The reaction of the
atoms to this energy depends on the charge of their shells. The
transition of energy produces an X-ray unique to the atomic
number and properties of a specific element [29]. Both XRPD and
SEM can be considered minimally destructive, as only a tiny
portion of powdered bone is required for acceptable results. It
should be noted that although SEM provides a useful tool for visual
assessment of a sample, the EDX feature provides a highly localised
and somewhat superficial analysis of sample chemistry. For this
reason, EDX elemental analysis is best used as a supplementary
tool to more advanced techniques such as XRPD.
Here we use XRPD, SEM and EDX analysis to examine and
compare the crystalline profiles of modern fresh bone, modern
burned bone and bone exposed to prolonged burial.
2. Methods
2.1. Sample selection and preparation
Thirty long bone samples were collected, including 21 from
three mature domesticated cow bones (Bos taurus), two from
archaic permafrost (20,000 years before present) bison bones
(Bison antiquus) and seven samples from domestic pigs (Sus scrofa)
(1 control sample, 1 sample per each of the 6 temperature
treatments). The three fresh bovine long bones and one fresh
porcine long bone were each cut into seven 1 cm thick cross-
sections. One section from each bone was retained as a control
sample and the remaining sections were burned in groups of four
for each of six temperature intervals (100
C, 300
C, 500
C, 700
C,
900
C, 1000
C). An electric kiln with in-built thermocouple
(Woodrow Hobby Fire Kiln Mini, © 2019 Keane Ceramics, NSW,
Australia) was programmed to maintain a consistent temperature
ramp-up speed of 400
C/h per run. Samples were introduced to
the kiln once the desired hold temperature was reached, burned
for 60 min, and then removed once the kiln completed the
cooldown cycle. Each sample was photographed and weighed
using a calibrated electronic scale before and after burning. Two
56
 M. Mckinnon et al. / Forensic Science International 313 (2020) 110328
Fig. 1. Illustration showing two key measurements used to establish characteristics (i.e. crystal size) of a target X-Ray Powder Diffraction (XRPD) peak: Full Width at Half
Maximum (FWHM) and 2-theta position (2u).
grams of powdered bone was obtained from each sample via
manual grinding with a mortar and pestle. Samples unable to be
easily pulverized using a mortar and pestle were subjected to an
additional slow speed drilling step (<100 revolutions per minute,
3.5 mm drill bit).
2.2. XRPD and SEM analysis
XRPD analysis was performed using a MiniFlex 600 X-Ray
Diffractometer 9 (©2019 Rigaku Corporation, Tokyo, Japan) with
CuKα radiation (l = 0.15418 Å) and a working voltage of 20 kV/2 mA.
Slit size was 1.25
u and wavelength was measured at 10
per
minute with a 0.02 step size. Peaks are only identified in XRPD if
they are larger than 3 nm and the crystallite in question makes up
at least 3% of the whole sample [16]. Once peak profiles were
obtained, phase analysis was performed using PDXL: Integrated
X-ray powder diffraction software (© 2019 Rigaku Corporation)
which uses the Crystallography Open Database (COD) as a source of
crystal structures and compositions. As apatite crystallites have a
hexagonal structure [6,30], peaks corresponding to the a- and
b-axis overlap substantially. As a result, average crystallite size can
only be obtained along planes perpendicular to the c-axis [31].
Based on published data, crystallographic planes (hkl/Miller
indices) found in HAp include 002, 211, 300, 202, 310, 222 and
213 [31,32], however some peaks may overlap at lower temper-
atures where samples are less crystalline. For this reason, only
planes clearly identified across all samples were used to establish
average crystallite size. The average crystallite size was calculated
for each plane using the Scherrer formula [33]:
D ¼  K l=ðbcosuÞ
Where D is the average crystallite size (in nm), K is the Debye-
Scherrer constant (0.89), l is the x-ray wavelength specific to the
machine used (0.15418 Å), u is Bragg diffraction angle and b is line
broadening in radians (peak full width at half maximum divided
by 2). Additionally, SEM analysis was performed using a FEI Quanta
450 FEG Environmental Scanning Electron Microscope (©2019
Thermo Fisher Scientific, Massachusetts, USA) with 15 kV and 3.0
spot size working conditions. Bone powder was applied to sticky
carbon buttons and sputter coated with platinum (10 nm) to
reduce interference and static discharge. This was performed by
Adelaide Microscopy staff using a Cressington 208 HR sputter
coater (©2019 Cressington Scientific Instruments, England, UK).
Captures were taken from three sites on each sample, at 20,000
and 50,000 magnification. At 50,000 magnification, images
were also assessed using EDX AZtec software (©2019 Oxford
Instruments plc, Nanoanalysis). In this software elemental
identification was performed for each sample, with platinum
excluded from analysis as sample coating would likely alter results.
Three point analyses and one area analysis were performed for
each sample site.
2.3. Statistical analysis
IBM SPSS [34] was used to test data distribution for normality
(Shapiro-Wilk) and the appropriate method for ANOVA and
multiple comparison testing (parametric or non-parametric).
Additionally, Spearman’s Rho correlation was used to explore
relationships between variables.
3. Results
3.1. Impact of sample size
The Shapiro-Wilk test showed the data was not normally
distributed. Given that with an appropriately large sample size
biological data almost always follows a normal distribution, it is
likely that our sample size caused this result rather than
abnormalities in the data itself. To adjust for this we applied
non-parametric analyses which are more robust than parametric
tests as they do not rely on distribution. This means that any
statistically significant results obtained in this way are made under
the most conservative conditions possible.
3.2. Post-burning sample changes
All samples exposed to sintering temperatures (>500
C) were
extremely brittle and easy to powder, however the fresh, low
temperature-treated and archaic samples required drilling. The
range of colour changes observed in post-burn samples can be
seen in Fig. 2. Sample weight was seen to decrease at all
temperatures but was more evident above 300
C. Porcine bones
57
 M. Mckinnon et al. / Forensic Science International 313 (2020) 110328
Fig. 2. Selection of bovine samples from three different individuals exposed to various temperatures. Above 700
C there was no further visible colour change.
showed a slightly greater weight loss than bovine samples
(Fig. 3).
3.3. XRPD profile analysis
The primary peaks evident in all x-ray diffraction patterns
were a positive match for archived phase data of HAp crystals
and were seen in all samples (Fig. 4.1-4.4). However, the high
intensity peak (32
2u) showed a substantial overlap between
211, 300 and 202 planes of reflection at lower temperatures, and
individual peaks were not apparent until 700
C. Only three
planes presented clearly defined peaks in all samples analysed.
These planes had Miller Indices (hkl) of 002, 310 and 213 and were
selected for further investigation. In samples burned at temper-
atures of 700
C and higher, a secondary phase was identified at
theta (2u ) positions 37
(200) and 53
(220) as CaO [35,36]
(Fig. 4.3, 4.4). Average HAp crystallite size values indicate that for
all crystallographic planes the greatest crystallite growth occurs
between the 500
C and 700
C (Table 1). Average crystallite size
Fig. 3. Histogram showing average percentage of bone weight loss for modern
bovine and porcine samples analysed within each temperature/treatment group
with standard error bars.
increased by a total of 191% (002), 805% (310) and 1260% (213)
from the lowest (unburned) to the highest (1000
C) temperature
groups.
3.4. XRPD ANOVAs and correlation analysis
Kruskal-Wallis H test (supplementary S1) showed significant
differences between temperatures for both HAp crystal size
(P < 0.013) and bone weight loss (P = 0.004). No significant differ-
ence was found in crystal size (P > 0.083) between unburned archaic
and modern bovine samples. Mann-Whitney tests showed in the
case of the 002 crystallite plane, the only significant change was seen
between 500 and 700
C where crystallite size increased dramati-
cally (36.9 nm). By comparison the 310 plane presented significant
changes at two intervals; a slight increase (1.46 nm) in crystallite
size between 100 and 300
C and a dramatic increase (55.75 nm)
between 500 and 700
C. Lastly, the 213 plane showed the most
change, with a small decrease (-4.38 nm) occurring from 100 to 300


C followed by increases from 300 to 500
C (6.33 nm), 500–700
C
(54.76 nm) and 900–1000
C (8.95 nm). Spearman’s rho non-
parametric correlation showed a significant (p < 0.001) positive
relationship between temperature and HAp crystallite size for all
three planes, as well as temperature and weight loss (Table 2).
3.5. SEM energy-dispersive x-ray spectroscopy descriptive statistics
Basic statistics (Table 3) show a trend towards increased carbon
and oxygen at higher temperatures, whilst calcium appears to
decrease. Other elements show peaks and drops at certain
temperature intervals, but as there is substantial overlap of
standard errors for most elements, no distinct pattern can be
discerned.
3.6. SEM ANOVAs and correlation analysis
Kruskal-Wallis H test (supplementary material S2) showed that
calcium was the only element that differed significantly between
temperature groups (P = 0.007), however no significant differences
in elemental composition were found between modern and
archaic samples (P > 0.237). Further post-hoc analysis showed
the weight percentage of calcium decreased significantly
(p = 0.016) from the unburned group to the highly burned
(1000
C) group. Spearman’s rho analysis (Table 4) showed a
relatively weak but still significant (p = 0.042) positive correlation
between temperature and carbon weight percentage. Correlation
58
 M. Mckinnon et al. / Forensic Science International 313 (2020) 110328

Fig. 4. X-ray diffraction phase patterns for archaic bison bone (4.1), modern untreated bovine bone (control) (4.2), and the same modern bone treated at 700
C (4.3) and
1000
C (4.4) temperature intervals.

Table 1
Average size (nm) of three primary HAp crystallographic planes in archaic samples and in modern samples before and after burning at controlled temperature intervals.

Temperature (
C) Avg. HAp Crystallite Size (nm)  SD Avg. HAp Crystallite Change (nm)

(002) (310) (213) (002) (310) (213)

Archaic: 15.18  4.72 7.44  0.53 2.98  0.79 – – –
Unburned: 19.69  1.11 7.16  0.52 5.26  0.44 – – –
100 17.46  2.51 6.11  1.10 8.06  1.14 2.23 1.05 2.8
300 18.67  3.33 7.57  0.73 3.68  2.45 1.21 *1.46 *4.38
500 17.81  0.41 7.04  0.40 10.01  1.03 0.86 0.53 *6.33
700 54.71  2.40 62.79  0.84 64.77  1.15 *36.9 *55.75 *54.76
900 54.62  6.22 59.48  4.88 62.60  2.78 0.09 3.31 2.17
1000 57.35  1.54 64.79  4.43 71.55  6.31 2.73 5.31 *8.95
* Significant (P < 0.05) Total Crystallite Increase 0-1000
C: 191% 805% 1260%

Table 2
Spearman’s rho correlation showing the strength of relationships between variables.

Crystallite Size (nm)/Bone Loss Crystallite Size/Temp Bone Loss/Temp

(002) (310) (213) (002) (310) (213)

Spearman's rho Correlation (r) **0.679 **0.710 **0.807 **0.718 **0.721 **0.845 **0.975
Sig. (2-tailed) 0.000 0.000 0.000 0.000 0.000 0.000 0.000
N 23 23 23 23 23 23 23
**
Correlation is significant at the 0.01 level (2-tailed).

Table 3
Average elemental weight percentage for samples within all treatment groups.

Carbon % Oxygen % Sodium % Magnesium % Calcium % Trace Elements %

Treatment Group: A.Bison 18.77 39.63 0.95 – 41.26 2.65

Unburned 13.00 36.51 1.02 0.72 52.36 –
100
C 18.76 44.07 1.19 0.66 36.38 2.21
300
C 26.92 45.70 1.33 0.74 32.79 –
500
C 24.12 41.23 0.99 0.76 35.93 –
700
C 19.03 42.55 0.90 0.96 35.72 1.49
900
C 20.07 48.43 1.16 0.82 30.62 –
1000
C 30.28 46.44 3.03 0.94 29.57 0.46

59
 M. Mckinnon et al. / Forensic Science International 313 (2020) 110328
Table 4
Spearman’s rho correlation showing the strength of relationships between elemental weight percentage and temperature increase.
Temp/Carbon
Spearman's rho Correlation (r) *0.418
Sig. (2-tailed) 0.042
N 24
*
Correlation is significant at the 0.01 level (2-tailed).
between temperature and calcium weight percentage was
similarly weak but significant (P = 0.000), however this relation-
ship was inverse.
3.7. Semi-quantitative analysis of SEM images
Fig. 5 shows a sample of SEM images comparing archaic bison
bone to modern bovine samples burned at various temperatures. A
notable similarity is apparent between the archaic bison samples
(Fig. 5a) and modern bovine samples burned at low temperatures
(Fig. 5c), with both presenting striated surfaces. At 500
C, burned
bone begins to restructure, with many samples developing small
pits (0.5 mm) on the surface (Fig. 5e). Samples burned at  700
C
showed the most dramatic change, with apatite crystals becoming
less dense and more granulated in appearance (Fig. 5f). From
600–1000
C these crystalline granules become visibly larger and
appear to “melt” into each other at the highest temperature
increment (Fig. 5h). Barring some minor individual sample
variation, analysis of the porcine SEM images showed a similar
pattern of crystallization from low to high temperatures (supple-
mentary material S3). The only exception to this was the samples
burned at 700
C, where granulation (though similar) presented
more ‘rod-shaped’ crystal aggregates compared to the relatively
spherical shaped aggregates of the bovine samples.
4. Discussion
In forensic research there is always a trade-off between
conducting studies realistically or accurately, and in this case
logistical restrictions led us to select the later. Electric furnaces
capable of reaching and holding controlled temperatures above
Fig. 5. SEM images taken at 20000 magnification (scale bar = 5 microns) – a. untreated archaic bison bone, b. untreated modern bovine bone and subsequent images show
the same modern bone treated at 100
C (c.), 300
C (d.), 500
C (e.), 700
C (f.), 900
C (g.) and 1000
C (h.).
Temp/Oxygen Temp/Sodium Temp/Phosphorous Temp/Calcium
0.191 0.113 0.170 *0.367
0.066 0.495 0.099 0.000
93 39 95 95
1000
C are extremely difficult to access as there is rarely a need for
them; finding one large enough to burn entire fleshed bones was not
possible. Herein lies one of the primary limitations of this study; to
suit the dimensions of the furnace we were forced to pre-section
bones prior to burning. This means we were not able to investigate
the likely impact of soft tissue shielding, and the effect of increased
bone size on heat penetration. Though these factors definitely
require consideration, measuring them was beyond the scope of
this study. The relatively small sample size must also be considered -
though significant results were obtained through reliable and
consistent analyses, increasing the sample size would only
strengthen these results and allow for further more in-depth
statistical models. Finally, the use of permafrost bones as repre-
sentatives of “archaic” samples should be discussed. Increased age
and severe diagenesis are not intrinsically linked, meaning that a
variety of factors besides age determine the degree of preservation.
Permafrost bones are a prime example of this, as they are far better
preserved than other samples originating from the same period.
Again this was an issue of logistics; due to the destructiveness of the
sampling process and the inherently precious nature of archaic
samples, it was not possible to access bones of non-permafrost
origin. Subject to increased availability, conducting similar analysis
on other types of archaic samples could prove highly beneficial in
supplementing the results presented here.
One of the primary directives of this study was to compare and
contrast changes to the mineralised structure of bone samples
exposed to either high temperatures or natural diagenesis. Given
the assumption that there is an intimate relationship between
bone hydroxyapatite and DNA, variation in mineral structure may
indicate that alterations to DNA extraction protocols are required
dependant on the taphonomic conditions to which the sample
60
 M. Mckinnon et al. / Forensic Science International 313 (2020) 110328
has been exposed. The link between DNA binding to HAp and DNA
preservation is not a new concept; in the past, HAp columns were
used for protein chromatography due to their affinity for DNA
binding [37]. Similarly, many studies have found DNA yield from
HAp containing structures such as bone or teeth is significantly
greater than that from other degraded tissues [38–40]. However,
it is only in the past decade or so that the relationship between
mineral structures and DNA preservation has been described as
more than simply “physical protection” of genetic material. This is
likely because the exact mechanisms of the relationship are
poorly understood, and its importance is often overlooked. It is
commonly stated that bone structure is the driving factor behind
DNA preservation, and denser cortical bone provides greater
physical protection for DNA than trabecular bone, where a greater
surface area is available for microbial invasion. This relationship is
multidirectional as increased microbial activity results in
decreased collagen and therefore increased porosity [5,20].
However, as far back as 1965, Bernardi suggested that the
interaction between nucleic acid and HAp calcium was the main
factor in DNA/HAp absorption [41]. Götherström et al. states that
HAp has a strong affinity for DNA and that ancient DNA
degradation can be linked to mineral resorption and crystallinity
loss [4]. Brundin et al. established that DNA from Fusobacterium
nucleatum incubated in a HAp medium produced significantly
greater quantities of nuclear material than samples incubated in
other media – this was attributed to DNA’s affinity for HAp binding
[42]. This affinity is believed to be a direct result of ionic binding
between positively charged calcium cations (Ca2+) in HAp, and the
negatively charged phosphate groups that form the sugar-
phosphate backbone of the DNA double helix [43,44]. With this
in mind, it is logical that dense bone would contain greater
quantities of DNA, not only as a result of a lack of porosity, but also
due to increased calcium content [41]. Therefore, any process
affecting the ratio and abundance of HAp cations and anions
(Ca2+/PO) can be expected to affect the quantity of DNA retained
within mineralised tissues. Exposure to high temperatures, as well
as natural weathering and diagenesis are destructive processes
that very likely alter these ionic ratios.
A mixture of amorphous material (likely collagen) and
crystalline HAp was seen in the XRPD phase patterns of the fresh
(and low temperature) modern bones examined in the current
study. This is consistent with Bartsiokas & Middleton’s statement
that “modern bone mineral can be regarded as poorly crystalline,
contaminated hydroxyapatite” [45]. Comparatively the archaic
samples examined in this study presented peak profiles and HAp
crystallites virtually indistinguishable from those of modern
unburned and low temperature (100
C) samples. This was not
consistent with the findings from other research demonstrating
that archaic samples produce sharper diffraction patterns indica-
tive of purer crystalline structures, as well as larger crystallites
than modern samples [46–48]. Visual assessment of SEM images
showed no evidence of the “rod-shaped crystals” Rogers et al.
described as being present in archaic bone and absent in modern
unburned samples [48]. Based on this study alone, archaic bison
bone could not be separated from fresh samples simply by using
XRPD, SEM and EDX. This suggests that diagenetic processes did
not greatly alter the mineral structure of the bone. However, when
cutting the archaic bone, we found it considerably harder to
section than the fresh and burned samples, indicating a difference
exists between the bone types. Although we were not able to
detect significant differences in the crystallinity of the permafrost
samples, there are likely other processes affecting the bone that
were not examined in this study. It is worth noting that although
permafrost samples are frequently used in ancient DNA studies,
they are far less degraded than bones of similar age found in other
regions. Heat treatment had a dramatic effect on the crystallinity of
samples. In most cases, the temperature that samples were
exposed to could be approximated based on crystal size and
appearance using SEM. In all samples exposed to 700–1000
C, HAp
crystals became significantly larger and CaO appeared as a
secondary phase bi-product. Concurrently, a significant decrease
in calcium content was observed with increased temperature.
Bones more abundant in Ca2+ ions have presented an increased
affinity for DNA absorption [37,49]. Knowing this, temperatures
high enough to break down and/or convert bone calcium to CaO
would likely result in a loss of any DNA bound to the calcium.
Alternately, lower temperatures likely have less of an effect on
HAp-bound DNA. Endogenous DNA bound to HAp potentially
remains present up until 500–700
C where significant recrystalli-
zation is observed. Although the sample size in this study is small,
the colour of the burned samples was clearly correlated with the
temperature to which they were burned [31]. If this is representa-
tive in all burned bones and is indicative of DNA content it would
provide an excellent screening tool for triaging of samples to
determine which are likely to provide genetic information.
With regard to inter-species differentiation; although HAp crystal
size was not deemed significantly different between porcine and
bovine samples, visual assessment of the SEM images showed
crystals of porcine samples heated to 700
C were more rod-shaped
than the plate-like crystals formed at the same temperature within
bovine samples. This suggests some morphological differences may
exist – or it could simply be that the rod-shaped crystals are
precursors to the next crystalline form, and also occur in bovine
samples at temperature ranges that were not recorded (i.e. during the
600
C range). Another explanation is simple topography; Rogers
et al. described bone mineral crystallites as having both rod and plate
like morphologies in most tissues reflective of local variation [48].
The only major difference seen between species was the percentage
of weight lost during burning, where porcine samples lost
significantly more weight than bovine samples. As bovine bones
are considerably larger than porcine bones this could indicate that
the deeper tissues in large bones are afforded more protection than
those of smaller bones, however due tothe limited numberof porcine
samples included in the current study this could also be a random
aberration. Although our results largely point towards a lack of
variation between the two species investigated, we cannot
automatically assume no difference will exist between non-human
(bovine and porcine) and human bones.
This work could be expanded on by including a broader range of
sample types - for example fresh and degraded modern human
(especially for forensic contexts), and severely degraded archaic
bone. Preservation of DNA depends on a range of variables, and
only changes in bone crystallinity and elemental composition were
analysed here. In addition to the techniques used in the presented
study, a variety of other techniques could be used to expand the
current understanding of how bone is altered by various forms of
trauma, and how this affects DNA recovery. This knowledge would
not only improve DNA recovery from challenging samples but
assist in the differentiation and possibly even identification of
fragmented remains.
Key points
1. Archaic bison bones and unburned modern bovine bones could
not be distinguished based on crystalline profiles and elemental
composition.
2. The crystalline profile of both bovine and porcine bone subjected
to heat remained unchanged until temperatures exceeded 500
C.
3. The crystalline profiles of burned bone suggests decalcification
prior to DNA extraction should not be performed on samples
that have not undergone recrystallization (600
C).
61
 M. Mckinnon et al. / Forensic Science International 313 (2020) 110328
4. To maximise success, different DNA extraction protocols are
required for differentially burned bone samples.
Funding
This work was supported by the University of Adelaide school of
Health and Medical Sciences.
Availability of data and material
The raw dataset generated for the purpose of this study is
available on reasonable request via the corresponding author.
CRediT authorship contribution statement
Meghan Mckinnon: Conceptualization, Methodology, Formal
analysis, Investigation, Data curation, Writing - original draft,
Writing - review & editing. Maciej Henneberg: Conceptualization,
Methodology, Writing - review & editing, Supervision, Project
administration, Funding acquisition. Ellie Simpson: Conceptuali-
zation, Methodology, Resources, Writing - review & editing,
Supervision, Project administration. Denice Higgins: Conceptuali-
zation, Methodology, Investigation, Resources, Writing - review &
editing, Supervision, Project administration, Funding acquisition.
Declaration of Competing Interest
None of the listed authors have conflicts of interest to declare.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at https://doi.org/10.1016/j.forsciint.2020.110328.
References
[1] Zhen Li, J. Pasteris, Chemistry of bone mineral, based on the hypermineralized
rostrum of the beaked whale Mesoplodon densirostris, Am. Mineral. 99 (4)
(2014) 645–653.
[2] C. Sosa, E. Vispe, C. Núñez, M. Baeta, Y. Casalod, et al., Association between
ancient bone preservation and DNA yield: a multidisciplinary approach, Am. J.
Phys. Anthropol. 151 (1) (2013) 102–109.
[3] C. Schwarz, R. Debruyne, M. Kuch, E. McNally, H. Schwarcz, New insights from
old bones: DNA preservation and degradation in permafrost preserved
mammoth remains, Nucleic Acids Res. 37 (10) (2009) 3215–3229.
[4] A. Götherström, M. Collins, A. Angerbjörn, K. Lidén, Bone preservation and DNA
amplification, Archaeometry 44 (3) (2002) 395–404.
[5] R. Hedges, Bone diagenesis: an overview of processes, Archaeometry 44 (3)
(2002) 319–328.
[6] S. Etok, E. Valsami-Jones, T. Wess, J. Hiller, C. Maxwell, et al., Structural and
chemical changes of thermally treated bone apatite’, J. Mater. Sci. 42 (23)
(2007) 9807–9816.
[7] G. Piga, A. Malgosa, T. Thompson, S. Enzo, A new calibration of the XRD
technique for the study of archaeological burned human remains, J. Archaeol.
Sci. 35 (8) (2008) 2171–2178.
[8] M. Harbeck, R. Schleuder, J. Schneider, I. Wiechmann, W. Schmahl, G. Grupe,
Research potential and limitations of trace analysis of cremated remains,
Forensic Sci. Int. 204 (1-3) (2011) 191–200.
[9] S. Ellingham, T. Thompson, M. Islam, The effect of soft tissue on temperature
estimation from burnt bone using fourier transform infared spectroscopy, J.
Forensic Sci. 61 (1) (2016) 153–159.
[10] M. Marques, A. Mamede, A. Vassalo, C. Makhoul, E. Cunha, D. Gonçalves, et al.,
Heat-induced bone diagenesis probed by vibrational spectroscopy, Sci. Rep. 8
(1) (2018)15935.
[11] T. Schwark, A. Heinrich, A. Preuße-Prange, N. von Wurmb-Schwark, Reliable
genetic identification of burnt human remains, Forensic Sci. Int. Genet. 5 (5)
(2011) 393–399.
[12] J. Jakubowska, A. Maciejewska, R. Pawlowski, Comparison of three methods of
DNA extraction from human bones with different degrees of degradation, Int. J.
Legal Med. 126 (2012) 173–178.
[13] D. Vanek, M. Silerova, V. Urbanova, L. Saskova, Dubska, et al., Genomic DNA
extraction protocols for bone samples: the comparison of Qiagen and Zymo
Research spin columns, Forensic Sci. Int. Genet. Suppl. Ser. 3 (1) (2011)
e397–e398.
[14] S. Seo, A. Zhang, H. Kim, J. Yi, H. Lee, et al., Technical note: efficiency of total
demineralization and ion-exchange column for DNA extraction from bone,
Am. J. Phys. Anthropol. 141 (2009) 158–162.
[15] O. Loreille, T. Diegoli, J. Irwin, M. Coble, T. Parsons, High efficiency DNA
extraction from bone by total demineralization, Forensic Sci. Int. Genet. 1 (2)
(2007) 191–195.
[16] J. Dissing, M. Kristinsdottir, C. Friis, On the elimination of extraneous DNA in
fossil human teeth with hypochlorite, J. Archaeol. Sci. 35 (6) (2008) 1445–
1452.
[17] N. Rohland, M. Hofreiter, Comparison and optimization of ancient DNA
extraction, Biotechniques 42 (3) (2007) 343–352.
[18] J. Davoren, D. Vanek, R. Konjhodzi c, J. Crews, E. Huffine, T. Parsons, Highly
effective DNA extraction method for nuclear short tandem repeat testing of
skeletal remains from mass graves, Croat. Med. J. 48 (4) (2007) 478.
[19] E. White, L. Hannus, Chemical weathering of bone in archaeological soils, Am.
Antiq. 48 (2) (1983) 316–322.
[20] N. Hoke, A. Rott, S. Johler, A. Reul, A. Beck, et al., How bone degradation, age,
and collagen extraction methods affect stable isotope analysis, Archaeol.
Anthropol. Sci. 7 (7) (2018) 1–18.
[21] L. Del Valle, O. Bertran, G. Chaves, G. Revilla-López, M. Rivas, DNA adsorbed on
hydroxyapatite surfaces, J. Mater. Chem. B 2 (40) (2014) 6953–6966.
[22] S. Bhattarai, S. Aryal, K.R. Bhattarai, N. Hwang, P, et al., Carbon nanotube-
hydroxyapatite nanocomposite for DNA complexation, Mater. Sci. Eng. C 28 (1)
(2008) 64–69.
[23] S. Ramesh, C. Tan, M. Hamdi, I. Sopyan, W. Teng, The influence of Ca/P ratio on
the properties of hydroxyapatite bioceramics, International Conference on
Smart Materials and Nanotechnology in Engineering, 6423(2007) , pp.
64233A.
[24] P. Bosch, I. Alemán, C. Moreno-Castilla, M. Botella, Boiled versus unboiled: a
study on Neolithic and contemporary human bones, J. Archaeol. Sci. 38 (10)
(2011) 2561–2570.
[25] R. Welker, Size analysis and identification of particles, in: R. Kohli, K. Mittal
(Eds.), Developments in Surface Contamination and Cleaning, William Andrew
Publishing, Norwich, NY, 2012, pp. 179–213.
[26] M. Figueiredo, A. Fernando, G. Martins, J. Freitas, F. Judas, et al., Effect of the
calcination temperature on the composition and microstructure of hydroxy-
apatite derived from human and animal bone, Ceram. Int. 36 (8) (2010) 2383–
2393.
[27] Y. Pang, X. Bao, Influence of temperature, ripening time and calcination on the
morphology and crystallinity of hydroxyapatite nanoparticles, J. Eur. Ceram.
Soc. 23 (10) (2003) 1697–1704.
[28] D. Ungureanu, N. Angelescu, R. Ion, E. Stoian, C. Rizescu, Synthesis and
characterization of hydroxyapatite nanopowders by chemical precipitation’
recent researches in communications, automation, signal processing, nanotech-
nology, Astron. Nucl. Phys. (2011) 296–301.
[29] Thermo Fisher Scientific, (2018) Accessed 23 September 2019, <https://www.
fei.com/introduction-to-electron-microscopy/stem/>.
[30] W. Landis, R. Jacquet, Association of calcium and phosphate ions with
collagen in the mineralization of vertebrate tissues, Calcif. Tissue Int. 93 (4)
(2013) 329–337.
[31] M. Greiner, A. Rodríguez-Navarro, M. Heinig, K. Mayer, B. Kocsis, Göhring, et al.,
Bone incineration: an experimental study on mineral structure, colour and
crystalline state, J. Archaeol. Sci. Rep. 25 (2019) 507–518.
[32] G. Poinern, S. Brundavanam, S. Tripathy, M. Suar, D. Fawcett, Kinetic and
adsorption behaviour of aqueous cadmium using a 30 nm hydroxyapatite
based powder synthesized via a combined ultrasound and microwave based
technique, Phys. Chem. 6 (1) (2016) 11–22.
[33] P. Scherrer, Bestimmung der Größe und der inneren Struktur von Kolloid-
teilchen mittels Röntgenstrahlen’, Nachrichten von der Gesellschaft der
Wissenschaften zu Göttingen, Math.-Phys. Klasse (1918) 98–100.
[34] IBM Corp. Released, IBM SPSS Statistics for Windows, Version 24.0, IBM Corp.,
Armonk, NY, 2016.
[35] S. *Bose, S. Saha, Synthesis and characterization of hydroxyapatite nano-
powders by emulsion technique, Chem. Mater. 15 (23) (2003) 4464–4469.
[36] H. *Bhagyalakshmi, M. Veerabhadraswamy, N. Venkatesha, Microwave
assisted synthesis of highly crystalline nano CaO from waste egg shells - a
low temperature record activity in glycerol carbonate, J. Ultra Chem. 14 (4)
(2018) 115–125.
[37] M. Okazaki, Y. Yoshida, S. Yamaguchi, M. Kaneno, J. Elliott, Affinity binding
phenomena of DNA onto apatite crystals’, Biomaterials 22 (18) (2001) 2459–
2464.
[38] D. Higgins, A. Rohrlach, J. Kaidonis, G. Townsend, J. Austin, Differential nuclear
and mitochondrial DNA preservation in post-mortem teeth with implications
for forensic and ancient DNA studies, PLoS One 10 (5) (2015) 1–17.
[39] P. Malaver, J. Yunis, Different dental tissues as source of DNA for human
identification in forensic cases, Croat. Med. J. 44 (3) (2003) 306–309.
[40] R. Gaytmenn, D. Sweet, Quantification of forensic DNA from various regions of
human teeth, J. Forensic Sci. 48 (3) (2003) 622–625.
[41] G. Bernardi, Chromatography of nucleic acids on hydroxyapatite columns,
Meth. Enzymol. 21 (1971) 95–139.
[42] M. Brundin, D. Figdor, G. Sundqvist, U. Sjögren, DNA binding to hydroxyapatite:
a potential mechanism for preservation of microbial DNA, J. Endod. 39 (2)
(2013) 211–216.
[43] E. Paget, L. Monrozier, P. Simonet, Adsorption of DNA on clay minerals:
protection against DNaseI and influence on gene transfer, FEMS Microbiol.
Lett. 97 (1-2) (1992) 31–39.
62
 M. Mckinnon et al. / Forensic Science International 313 (2020) 110328
[44] B. Busse, M. Hahn, M. Soltau, J. Zustin, K. Püschel, G. Duda, et al., Increased
calcium content and inhomogeneity of mineralization render bone toughness
in osteoporosis: mineralization, morphology and biomechanics of human
single trabeculae, BONE 45 (6) (2009) 1034–1043.
[45] A. Bartsiokas, A. Middleton, Characterization and dating of recent and fossil
bone by X-ray diffraction, J. Archaeol. Sci. 19 (1) (1992) 63–72.
[46] E. Stathopoulou, V. Psycharis, G. Chryssikos, V. Gionis, G. Theodorou, Bone
diagenesis: new data from infrared spectroscopy and X-ray diffraction,
Palaeogeogr. Palaeoclimatol. Palaeoecol. 266 (3-4) (2008) 168–174.
[47] D. Misra, Surface chemistry of bone and tooth mineral, Methods Calcified
Tissue Prep. (1984) 435–465.
[48] K. Rogers, S. Beckett, S. Kuhn, A. Chamberlain, J. Clement, Contrasting the
crystallinity indicators of heated and diagenetically altered bone mineral,
Palaeogeogr. Palaeoclimatol. Palaeoecol. 296 (1-2) (2010) 125–129.
[49] M. Jordan, A. Schallhorn, F. Wurm, Transfecting mammalian cells: optimiza-
tion of critical parameters affecting calcium-phosphate precipitate formation,
Nucleic Acids Res. 24 (4) (1996) 596–601.
63
 Chapter 5: Effects of Thermal Insult on

Bone Tissue as Observed by Micro

Computed Tomography

Published: Mckinnon, M, Henneberg, M, Simpson, S & Higgins, D 2021, ‘Effects of thermal insult on

bone tissue as observed by micro computed tomography’, Forensic Imaging, vol. 24, doi:

https://doi.org/10.1016/j.fri.2021.200437

64
 Statement of Authorship
Title of Paper: Effects of thermal insult on bone tissue as observed by micro computed tomography
Publication Status: ☒ Published

☐ Accepted for Publication

☐ Submitted for Publication

☐ Unpublished & Unsubmitted work written in Manuscript Style

Publication Details: Published in Forensic Imaging in February 2021.

Principal Author (Candidate): Meghan Mckinnon

Contribution to the paper: Conceptualization, methodology, formal analysis, investigation, writing
– original draft, writing – editing
Overall percentage (%): 85%
Certification: This paper reports on original research I conducted during the period of my Higher
Degree by Research candidature and is not subject to any obligations or contractual
agreements with a third party that would constrain its inclusion in this thesis. I am
the primary author of this paper.

23.06.21
Signature & Date: ____ ____________________________________

Co-Author Contributions
By signing the Statement of Authorship, each author certifies that:
i. the candidate’s stated contribution to the publication is accurate (as detailed above);
ii. permission is granted for the candidate in include the publication in the thesis; and
iii. the sum of all co-author contributions is equal to 100% less the candidate’s stated contribution.
Co-Author: Maciej Henneberg
Contribution to the paper: Conceptualization, methodology, writing – review & editing,
supervision, project administration, funding acquisition.

Signature & Date: ____ 23.06.21

________________________________
Co-Author: Ellie Simpson
Contribution to the paper: Conceptualization, methodology, writing – review & editing,
supervision, project administration.

Signature & Date: __ 23.06.21

_________________________________________
Co-Author: Denice Higgins (secondary author)
Contribution to the paper: Conceptualization, methodology, investigation, resources, writing –
review & editing, supervision, project administration, funding acquisition.

Signature & Date: ____ 23.06.21

_______________________________________

65
 Forensic Imaging 24 (2021) 200437

Contents lists available at ScienceDirect

Forensic Imaging
journal homepage: www.sciencedirect.com/journal/forensic-imaging

Invited article

Effects of thermal insult on bone tissue as observed by micro

computed tomography
Meghan Mckinnon a, *, Maciej Henneberg a, Ellie Simpson b, Denice Higgins c
a
Discipline of Anatomy and Pathology, Adelaide Medical School, the University of Adelaide, Helen Mayo North, Frome Road, Adelaide, SA 5000, Australia
b
Forensic Science SA, Adelaide, Australia
c
Forensic Odontology Unit, Adelaide Dental School, the University of Adelaide, Adelaide, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: A thorough understanding of the effects of heat on bone morphology is required for reliable analysis of burnt
Burning skeletal remains in forensic anthropology and bioarchaeology. However, thermal damage often prevents appli­
Bone porosity cation of the usual analytical techniques in both modern and archaeological contexts. This study uses Micro
Bone microstructure
Computed Tomography (MicroCT) to assess heat-induced changes in incinerated bovine (Bos taurus) long bones
Micro computed tomography
at temperatures between 100 and 700 ◦ C. Scans provided data on bone porosity and surface to volume ratio,
which were compared between samples. Macroscopic comparisons were also undertaken using 3D re­
constructions of samples. Results showed a decrease in porosity between unburned samples and those burned at
the highest temperature (700 ◦ C). This change was not unidirectional, with an unexpected increase at 500 ◦ C.
The lack of uniform exponential change suggests that more than one mechanism influences bone morphology in
response to thermal insult. The initial significant decrease in porosity may be attributed to the loss of organic
material (100–200 ◦ C) which had the most dramatic impact on bone morphology, whereas the increase at 500◦ C
coincides with hydroxyapatite re-crystallisation which is known to occur around this temperature. This study
demonstrates that MicroCT can be readily applied to burned bone which is often fragile and difficult to examine.
Additionally, unlike most microscopic techniques MicroCT is three dimensional, allowing the internal structure
of the entire bone to be investigated without destruction of the sample. The changes observed in bone
morphology due to heating provide valuable insight which can inform subsequent investigation and analysis of
these samples.

Introduction
Skeletal tissues persist long after soft tissues have been lost and can
provide insight into the identity of an individual, their lifestyle and even
how they died. This is valuable in many contexts including bio­
archaeological, anthropological and forensic investigations. Thermal
insult, however, can significantly affect the structural integrity of bone,
obscuring signs of antemortem changes and limiting the analytical
techniques available to obtain data. For example, radiocarbon dating,
and stable isotope analysis of ancient bone are routinely used to
reconstruct past diets, environments, and migration patterns [1, 2].
However, following a) loss of collagen, b) bone recrystallization (i.e.
after burning or prolonged natural diagenesis), or c) exchange/fixation
of structural carbonate, these analyses are not always possible [3].
In addition to chemical and microstructural alterations, bone can
severely warp under intense heat. This can prove challenging for an­
thropologists who rely primarily on gross morphology to determine
biological characteristics such as origin (i.e. human or non-human), sex,
height and age [4]. Evidence of trauma (e.g. blunt force, knife wounds)
may also be lost due to heat induced changes, depending on the severity
and duration of heating. In addition to fracturing and shrinkage, a va­
riety of features (Fig. 8) such as colour, water content, collagen struc­
ture, hydroxyapatite crystallization and the appearance of calcium oxide
are known to be affected when bone is exposed to high temperatures [5,
7]. The exact temperature that these changes occur at varies depending
on factors such as bone density, burn duration, and calcium phosphate
ratio [6, 7]. Blackened bone indicates a loss of the majority of organic
material, that changes to elemental carbon. Whitened bone indicates
complete replacement of organic carbon with hydroxyl groups, and
conversion from an amorphous to crystalline mineral [8]. It is

* Corresponding author.
E-mail address: meghan.mckinnon@adelaide.edu.au (M. Mckinnon).

https://doi.org/10.1016/j.fri.2021.200437
Received 27 January 2021; Accepted 3 February 2021
Available online 6 February 2021
2666-2256/© 2021 Elsevier Ltd. All rights reserved.

66
M. Mckinnon et al.
notoriously difficult to identify at precisely what temperatures these
changes occur due to the large number of variables; bone type and
density (e.g. cortical bone vs. trabecular) being a primary factor.
Microscopic analysis is often used to examine bone to reveal iden­
tifying characteristics and antemortem and postmortem changes. How­
ever, traditional microscopy techniques are destructive and only provide
a localised view of a small area, hence, variation in other regions of the
bone may be overlooked. Additionally, both diagenesis and thermal
insult can severely degrade bone tissue, limiting histomorphology and
microscopic analysis [9]. Decalcifying samples for slicing and staining
becomes nearly impossible for incinerated bone due to the loss of
organic matter and collagen framework during burning. A study by
Cattaneo et al. [10] found that despite significant shrinkage of osteons,
microscopy was highly accurate in identifying the origin (human or
non-human) of burned bone samples embedded in resin. However,
scanning electron microscopy of resin embedded bone still only provides
a localised view of a single surface of the sample, requires specialised
processing and causes significant destruction of the bone. The diffi­
culties associated with investigation of burned samples indicate a need
for a visual triaging tool that would allow non-destructive assessment of
bone for targeted analysis. MicroCT could potentially fill this niche, as it
allows three-dimensional inspection of an entire sample with the ability
to slice through at various levels without physically slicing the bone.
Computed Tomography (CT) is an imaging tool common in medical
and bioarchaeological fields that is now also used in forensic in­
vestigations [11–13]. Being non-invasive, CT is ideal for this type of
work, where practical and/or ethical boundaries prevent the destruction
of delicate samples. In some cases, CT (along with Magnetic Resonance
Imaging) is beginning to take the place of full autopsies, with cause of
death established from in-depth examination of full-body scans [14, 15].
By comparison, Microcomputed Tomography (MicroCT) is still in its
infancy when it comes to industry applications and is primarily used for
research. Although much higher resolution than ordinary CT, MicroCT is
substantially more expensive, and might be considered inutile as
currently only small samples can be analysed. However, where a
microscopic approach is required e.g. fragmented/small samples,
MicroCT could well augment histological analyses and/or be used as a
screening tool. Kuhn et al. [16] demonstrated the usefulness of this
technique on fresh human bone by comparing trabecular bone cube
micro-CT scans to corresponding histological sections, finding that
micro-CT measures of bone volume fraction (PP) and trabecular plate
density (PL) were not significantly different from those measured from
histological sections. To date Sandholzer and collaborators [17, 18]
appear to be the only researchers consistently using MicroCT to inves­
tigate heat induced changes in biological material. Thermal insult can
severely hinder the process of identification, in both forensic and
anthropological contexts [19–21]. In order to establish an accurate
biological profile, anthropologists need a thorough understanding of the
changes occurring in bone during the processes of burning. This current
study aims to examine these changes in bone burned at various tem­
peratures as observed by MicroCT.
Methods
Samples
Eight 10mm thick cross-sections were collected from the cortical
bone shaft of three individual domestic cow (Bos taurus) femurs, with
each allocated to a different temperature treatment (three sections (one
from each bone) per temperature group, N=24). An electric kiln with
digital temperature control and in-built thermocouple (Woodrow Hobby
Fire Kiln Mini, © 2019 Keane Ceramics, NSW, Australia) was used to
burn samples at 100 ◦ C temperature intervals (unburned, 100 ◦ C,
200 ◦ C, 300 ◦ C, 400 ◦ C, 500 ◦ C, 600 ◦ C, 700 ◦ C). For consistency a ramp-
up (heat-up) speed of 400 ◦ C/h per run was maintained, and upon
reaching the desired temperature samples were inserted and burned for
Forensic Imaging 24 (2021) 200437
60 min durations. After 60 min elapsed, a cooldown cycle would begin
and samples could be removed.
Scanning & reconstruction
In order to develop the correct CT protocols for the samples, a pilot
study of three samples burned at low (unburned) intermediate (400 ◦ C)
and high (700 ◦ C) temperatures was conducted. A Bruker SkyscanTM
1276 apparatus and associated software was used. By this process the
following settings for the study were determined:
• Filter: aluminium-copper (Al-Cu)
• Voxel resolution: 8.04 µm
• Exposures: 820 ms
• Voltage: 100 kV
• Current: 200 MA
• Rotation step (degrees): 0.200
Again, we used test samples to determine the optimal settings for
reconstruction, using NRecon (Skyscan, Kontich, Belgium). Optimal
settings were determined as:
• Smoothing: 2
• Ring artefact reduction: 8
• Beam hardening correction: 30 %
• Output log: 0.029
Using the Skyscanner’s native software CTAn (Skyscan, Kontich,
Belgium) we conducted both 2D and 3D analysis on a consistent volume
of interest (VOI) containing 600 slices for each sample. At this point one
of the most vital factors of MicroCT analysis had to be considered:
thresholding. All morphometric variable measurements can only be
performed on segmented or binarised images, which is achieved via
“thresholding”. The threshold is a number corresponding to a greyscale
value above which all pixels are considered bone, and below which are
considered non-bone [22, 23]. One method of obtaining these values is
“local” thresholding, where values are determined only by the grey
levels of surrounding voxels. This provides a uniform threshold that is
appropriate for samples that possess the same mineral structure and
density but may not suit samples presenting local variation i.e. cortical
and trabecular bone. An alternative method used to overcome the issue
of local variation is “adaptive” thresholding, whereby multiple local
threshold values are collected, and the average of the maximum and
minimum values are used to binarise the data [22, 16]. Due to the
extreme likelihood of morphological variation in burned bone, we opted
to use the latter method. Adaptive pre-thresholding within the 2D space
was trialled on various samples until 50-140 (with a radius of 6 voxels)
was determined the most universally appropriate threshold setting.
Following thresholding, scans were despeckled (pixels <250
removed from 2D space), and a shrink-wrap function applied to restrict
further analysis to the targeted area. Finally, 3D analysis was used to
generate quantitative data on the following variables: bone surface to
bone volume ratio (BS/BV), mean pore volume (Po.V) and mean pore
surface area (Po.S), bone (BV) and porosity (Po) percentage, and surface
to volume ratio of pores (PS/PV). Additionally, 3D modelling was used
to visually display sample changes, and inverse thresholding was
applied to highlight regions of non-bone (porosity) within the bounds of
each sample. This was used to create a visual reconstruction of porosity.
A summary of variables and their abbreviations is provided in Table 1.
These variables are commonly used for measuring changes in bone
morphology in clinical studies, and are likely to be severely altered by
thermal insult. The information in the table was based off the Bruker
MicroCT handbook [24].
In simplified terms, bone surface to volume ratio (BS/BV) measures
the relationship between the whole exterior surface of the bone
compared to the volume of solid material within the bounds of the bone.
67
M. Mckinnon et al. Forensic Imaging 24 (2021) 200437

Table 1 Table 2
Description, definition and abbreviation of all MicroCT measured variables. Average values obtained for each variable assessed at each temperature interval.
Variable names follow Bone ASBMR nomenclature [25]. BS/BV Po.V Po.S BV Po PS/ N
Variable Definition Abbreviation Unit (mm-1) (mm3) (mm2) (%) (%) PV

Bone surface / bone The ratio of solid surface to BS/BV mm- Unburned 18.84 55.48 2,355 73 27 52.33 3
volume ratio volume measured in 3D within 1 100◦ C 14.71 28.04 972 76 24 45.33 3
the VOI (Volume Of Interest). 200◦ C 10.11 14.46 511 80 20 41.33 3
Total pore space The total volume of all pores Po.V mm3 300◦ C 6.00 5.13 325 94 6 88.33 3
volume within the VOI. 400◦ C 12.17 10.50 624 86 14 75.00 3
Total surface area of The total surface area of all Po.S mm2 500◦ C 17.31 15.48 881 80 20 63.33 3
pores pores within the VOI. 600◦ C 12.40 10.12 613 86 14 76.00 3
Percent bone volume The proportion of the VOI BV % 700◦ C 7.79 5.46 303 90 10 76.67 3
(bone volume/total occupied by binarised solid
volume) objects.
Percent pore volume Total porosity is the volume of Po % however, the values for unburned samples are significantly greater than
(pore volume/total all pores as a percent of the total for all temperature treatments, despite the increase at 500 ◦ C (Table 2).
volume) VOI volume.
Comparatively, change in the percentage of bone and porosity (BV and
Pore surface area to Average ratio of pore surface PS/PV mm-
volume ratio (pore area to pore volume within the 1 Po) only somewhat follows this trend, and the extremely low porosity at
surface/pore total VOI. 300 ◦ C is more prominent than the subsequent increase at 500 ◦ C
volume) (Table 2). Pore surface to pore volume ratio (PS/PV) showed the greatest
deviation from this general trend, presenting lower values in unburned,
100 ◦ C 200 ◦ C samples before increasing at 300 ◦ C (Fig. 3).
This can be indicative of the surface complexity of the sample, for
Data shown in Table 3 indicate significant differences only exist in
example a less uniform surface would present a higher surface area to
Po.V (P=0.006), Po.S (P=0.003) and PS/PV (P=0.001) at different
volume ratio, compared to a flatter, more uniform surface. This also
temperature intervals. Furthermore, post-hoc analysis showed that Po.V
applies to the complexity of pores (PS/PV). Total pore surface (Po.S) and
was not significantly different between unburned and 100 ◦ C samples
volume (Po.V) both measure pores contained within the internal bounds
(P=0.281), however, unburned samples were significantly more porous
of the solid object. These may be defined circular pores, cavities or
than samples burned over 200 ◦ C (P<0.034). This was also the case for
fractures, the consistency of which influences the average surface to
Po.S, where unburned samples were not significantly different from
volume ratio of pores. Again if this value is high it can indicate the
100 ◦ C samples (P=0.059), but pore surface area significantly decreased
presence of multifaceted and/or varied porosity, whereas low values
over 200◦ C (P<0.007). Comparatively, PS:PV multiple comparisons
indicate more uniform porosity and/or decreased surfaces e.g. rounded
showed that samples burned at 200 ◦ C degrees had significantly lower
pores.
ratio of pore surface to pore volume than those of higher temperatures
(with the exception of 500 ◦ C samples). Samples burned at 300 ◦ C
Scanning electron microscopy
presented ratios that were significantly higher than those of lower
temperatures (<200 ◦ C).
Following CT scanning, a selection of samples were mounted on
Pearson’s correlation showed that all variables with the exception of
carbon tabs and coated by Adelaide Microscopy staff for Scanning
BS/BV were significantly, but weakly correlated (P<0.01, R<0.52) with
Electron Microscopy (SEM). A 10nm thick platinum coating was applied
temperature (Table 4). Of these significant correlations, only BV and PS/
using a Cressington 208 HR sputter coater (©2019 Cressington Scientific
PV was positive (bone volume and pore surface to volume ratio
Instruments, England, UK), samples were then examined at various
increased with increasing temperatures), whilst the other variables were
magnifications ranging from <200x (mm) - 20000x (µm). A FEI Quanta
all negatively correlated (decreased with increasing temperatures). The
450 FEG Environmental Scanning Electron Microscope (©2019 Thermo
strongest correlation exists between increasing temperature and
Fisher Scientific, Massachusetts, USA) with a working voltage of 15 kV
decreasing Po.V (R=0.72). Logarithmic curves fitted to the data (Figs. 4
and 3.0 spot size was used to collect captures of bone topography for
and 5) showed the same significance (or lack of) for each variable, with
visual analysis.
increasing temperature and decreasing Po.V showing the best fit for this
regression.
Statistical analysis

Data collected in this study followed a normal distribution, therefore
parametric analyses were justified. One-way ANOVAs were conducted
to determine if significant differences exist between temperature treat­
ments, and Tukey’s post-hoc tests were then performed to establish
variation between specific groups. Pearson’s correlation was applied to
show the relationship (or lack thereof) between bone morphometry
variables and temperature change, and if said relationships were pre­
sent, raw data were plotted and displayed using a logarithmic regression
model.
Results
Although not always significant, an overall trend was apparent for all
variables investigated, with the possible exception of BV and PS/PV.
From initially high surface and porosity values (Table 2) of unburned
bone, there was a dramatic decrease until 300 ◦ C then a linear increase
to a peak at 500 ◦ C degrees, before decreasing once more. This is
emphasised in Fig. 1, where BS/BV of the unburned and 500 ◦ C sample
groups clearly overlap. This trend is again apparent in Fig. 2a and b,
Visual assessment
3D cross-sections of samples showing significant or remarkable
variation in all variables measured are presented in Fig. 6. Other samples
are shown in supplementary material 1. The presence of microfractures
was observed in high temperature samples, however, these were mostly
surface level, and did not show up clearly on the 3D models. The un­
burned and 700 ◦ C samples showed the greatest difference, particularly
in porosity. Before burning, bone presented large circular pores, which
decreased in size during burning. Narrow porous “channels” appeared at
higher temperatures. Some temperatures (e.g. 300 ◦ C) showed the
presence of more uniform porosity, which may represent fractures along
consistent planes. At 700 ◦ C nearly no porous space was present. This
finding was mirrored in the SEM images collected (Fig. 7). Again only
samples showing remarkable variation are included here, more can be
found in supplementary material 2. As with the 3D reconstruction and
analysis, 300 ◦ C, 500 ◦ C and 700 ◦ C samples presented substantially
different surface appearances. Micro fracturing appears at 300 ◦ C (Fig. 7
A and D) with very few small/circular pores, whilst at 500 ◦ C (Fig. 7 B

68
M. Mckinnon et al. Forensic Imaging 24 (2021) 200437

Fig. 1. Histogram of mean bone surface to bone volume ratio (BS/BV) across increasing temperature intervals.

2a. 2b.

Fig. 2. Histograms showing a) mean pore surface (Po.S), and b) mean volume of pores (Po.V) across increasing temperature intervals.

Fig. 3. Histogram showing the ratio of pore surface to pore volume (PS/PV).

69
M. Mckinnon et al. Forensic Imaging 24 (2021) 200437

Table 3 complexity changes under thermal insult. The primary finding was that
One-way ANOVA F-test table showing variation within each morphometric bone porosity significantly decreased after 100 ◦ C, whereas the ratio of
variable across all temperature treatments. pore surface to pore volume was significantly higher at temperatures
Variable x Temperature F Sig. above 300 ◦ C. The increased ratios suggest that pores are changing in
BS/BV (mm-1) 1.657 0.190
shape and complexity at higher temperatures, which was supported by
Po.V (mm3) 4.487 0.006* visual assessment. Using both MicroCT reconstruction software and SEM
Po.S (mm2) 5.236 0.003* topographical microscopy, dramatic changes in general bone structure
BV 2.311 0.078 from unburned to incinerated (700 ◦ C) samples were observed. Focusing
Po (%) 2.311 0.078
specifically on porosity, only the 3D MicroCT reconstructions clearly
PS/PV 6.979 0.001*
show the change in pore dimensions at 300 ◦ C, as well as the overall loss
*
=P is significant of porosity at high temperatures.
It is difficult to directly compare results obtained in the current study
to other research, as we could find no examples of MicroCT being used to
Table 4
measure porosity of incinerated bone. However, the observed overall
Pearson’s correlation (one-tailed) and curve fitting (logarithmic regression)
decrease in pore space was unsurprising, based on a broader under­
between temperature and morphological variables.
standing of the taphonomy of burned bones. Many studies using visual
Pearson’s Correlation Logarithmic Regression
assessment (e.g. SEM, radiography, histology) attribute loss of porosity
Variable x Temperature N Sig. (1-tailed) R R2 Sig.
to a) major degradation of organic material (primarily water and
BS/BV (mm-1) 24 0.127 0.32 0.10 0.118 collagen), and b) recrystallization of bone mineral. Firstly organic ma­
Po.V (mm3) 24 0.001* 0.72 0.51 0.000*
terial is lost at temperatures above 100 ◦ C and below 200 ◦ C [26], with
Po.S (mm2) 24 0.006* 0.65 0.43 0.000*
BV 24 0.014* 0.52 0.27 0.009* collagen denaturation occurring at around 155 ◦ C [8]. This process re­
Po (%) 24 0.014* 0.52 0.27 0.009* sults in decreased porosity as the bone loses elasticity and shrinks, and
PS/PV 24 0.002* 0.58 0.34 0.002* cavities previously containing organic material are filled with carbon­
*=Correlation is significant ised debris [8, 27]. Secondly, carbon is replaced by hydroxyl groups
which, in the case of dense cortical bone, fill in the spaces left by organic
and E) samples resembled those of lower temperatures (unburned- material resulting in a purer, crystalline form of hydroxyapatite
100 ◦ C). The most dramatic change was seen in the 700 ◦ C sample (Fig. 7 [28–30]. The greatest decrease in porosity observed was between un­
C and F), where small, circular pores appear consistently alongside burned samples compared to 100 ◦ C samples. Although this could be a
larger, narrow crescent-like pores. result of the aforementioned loss of organic material, it is surprising that
this decrease was greater than in high temperature samples (i.e.
Discussion 500–700 ◦ C) where significant structural changes were also seen. One
possible explanation for this is the trace amounts of trabecular bone seen
The purpose of this study was twofold; to provide insight into the in some 3D reconstructions (particularly unburned samples), which
structural changes occurring in bone during burning, and to explore could have resulted in higher porosity values. Despite being collected
MicroCT as an investigative tool for burned bone compared to other from the same section of femoral bone, it was impossible to generate
techniques (i.e. SEM). As a non-destructive method capable of three identical sample slices. Although we did not visually observe
dimensional analysis, MicroCT is ideal for fragile samples such as pre-treatment differences from one sample to the next, it is not unlikely
incinerated bone. We aimed to provide a better understanding of what that some degree of microscopic/elemental disparity existed prior to
specific changes could be observed using this method, and highlight the burning. As samples were collected from femoral shafts and no trabec­
benefits it may have over other techniques. A major advantage of ular material was apparent macroscopically, it was surprising to see it on
MicroCT is the ability to supplement quantitative data with visualised the MicroCT images. Less trabecular bone was apparent in samples
3D reconstructions. Bone porosity can be assessed numerically as well as burned at temperatures over 100 ◦ C. The fragile lattice structure of
visually, providing a more cohesive understanding of how pore trabecular bone is less likely to survive carbonisation, meaning it is often

Fig. 4. Logarithmic regression model showing a) mean surface of pores (Po.S) and b) mean volume of pores (Po.V) with increasing temperature.

70
M. Mckinnon et al. Forensic Imaging 24 (2021) 200437

Fig. 5. Logarithmic regression model showing a) the change in percentage of bone (BV) and porosity (Po) and b) ratio of pore surface to volume (PS/PV) with
increasing temperature.

Fig. 6. MicroCT 3D cross sections of samples at selected temperature intervals. Bone material is shown in greyscale. The second row shows an “inverse” repre­
sentation of bone, with non-bone areas in the region of interest selected to represent porosity.

lost before recrystallization occurs – this is consistent with our findings
[31–33]. Some research has shown that at lower temperatures (<
250 ◦ C) the circularly organised lamellar layers of Haversian systems
undergo substantial changes, primarily the broadening of Haversian
canals as observed with SEM [34]. This reorganisation of bone internal
structure, along with the dramatic recrystallization of hydroxyapatite at
between 500 and 600 ◦ C [35–37] are likely a primary drivers of
changing porosity.
It is apparent from this work that MicroCT may have potential ap­
plications in a much greater range of areas than it is currently used in.
Thompson [38] noted that heat-induced changes in porosity and frac­
turing adversely affected standard anthropological techniques such as
morphological and metric assessment of bone. MicroCT could provide a
means of quantifying how much the internal structure of bone may have
been altered during burning, and digital measurements can be taken
from CT scans that account for bone shrinkage and degradation. Another
potential application is non-invasive investigation of bone pathologies,
for example porous lesions on the orbital roof (cribra orbitalia), growth
arrest lines of varying bone densities (Harris lines), and thickening of the
frontal bone resulting from hormonal changes (hyperostosis frontalis)
[39–41]. While these morphological changes may be clearly identified
on unburned skeletal remains, incineration can make macroscopic
diagnosis impossible from examination of bone exterior alone [42]. It
has been postulated that heat-induced changes in bone structure and
composition may be similar to that of diagenetically altered ancient
bone [43]. This has largely been assessed using x-ray powder diffraction
(XRPD) analysis, which calculates changes in mineral crystallite size
though measurement of diffraction peaks [44, 45]. Although a useful
technique, XRPD requires some destruction of the sample, and does not
allow visualisation of the interior microstructure of bone. MicroCT could
augment this process and provide some contextual information on the
similarities (or lack thereof) between modern incinerated and ancient
bones. Aside from a lack of research, the major issue with implementing
MicroCT in many of the aforementioned contexts comes down to
logistical and financial constraints. Like all novel technologies MicroCT
will likely became more affordable with time. Most current MicroCT

71
M. Mckinnon et al. Forensic Imaging 24 (2021) 200437

Fig. 7. SEM images of samples from three different temperature ranges (300, 500 and 700). These are presented at lower magnification (< 150x) to show the uniform
structure of the bone (A-C), as well as higher magnification (600x) for a more localised view (D-F).

Fig. 8. Simplified diagram showing changes known to occur in bone at increasing temperatures. Changes in bone colouration are represented above each tem­
perature interval, however it must be noted that this (and other factors shown) is variable. Created with BioRender.com.

systems are excellent for investigating small, localised sections of bone.
However, a major issue is sample housing volume, meaning larger intact
samples (i.e. an entire skull) can only be digitised using ordinary CT
instead of the high resolutions afforded by MicroCT. Large volume
MicroCT scanners do exist, for example the Nikon XT H 225ST CT
cabinet system, which can scan large and heavy objects at much higher
resolutions than medical CT systems. However, these are uncommon (at
the time of writing there is only one in Australia), and the standard
resolution is still less than that of a small volume MicroCT system such as
the Bruker SkyscanTM 1276. If large volume MicroCT scanning could be
applied to a range of incinerated intact bones, in addition to intact bones
affected by various pathologies, a more complete understanding of the
similarities (or differences) in porosity could be assessed.
Conclusion
This study showed that changes to bone microstructure, specifically
porosity, could be assessed using MicroCT in a more efficient and
possibly more effective manner than SEM. Although it is highly unlikely
that MicroCT will entirely replace the need for other analytical

72
M. Mckinnon et al.
techniques, its value as a potential triaging tool should be considered.
Due to the three dimensional nature of CT scanning, the internal
structure of precious or delicate samples can be investigated without
destruction. This could allow for identification of artefacts or regions of
interest that could subsequently be avoided or targeted for further
investigation. The application of MicroCT in casework certainly war­
rants further study.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Funding
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
Supplementary materials
Supplementary material associated with this article can be found, in
the online version, at doi:10.1016/j.fri.2021.200437.
References
[1] C Sullivan, H Krueger, Carbon isotope analysis of separate chemical phases in
modern and fossil bone, Nature 292 (5821) (1981) 333–335.
[2] D Pate, R Henneberg, M Henneberg, Stable carbon and nitrogen isotope evidence
for dietary variability at ancient Pompeii, Italy, Mediterr. Archaeol. Archaeom. 16
(1) (2016) 127–133.
[3] S Ambrose, J Krigbaum, Bone chemistry and bioarchaeology, J. Anthropol.
Archaeol. 22 (3) (2003) 193–199.
[4] D Gonçalves, Cremains. The Value of Quantitative Analysis for the
Bioanthropological Research of Burned Human Skeletal Remains, University of
Coimbra., 2012.
[5] D Ubelaker, The forensic evaluation of burned skeletal remains: a synthesis, Foren.
Sci. Int. 183 (1-3) (2009) 1–5.
[6] W Brooks, C Galvez Mora, J Jackson, J McGeehin, D Hood, Coal and cremation at
the Tschudi burn, Chan Chan, northern Peru, Archaeometry 50 (3) (2008)
495–515.
[7] B Herrmann, G Grupe, Trace element content in prehistoric cremated human
remains, in: G Grupe, B Herrmann (Eds.), Trace Elements in Environmental
History, Springer-Verlag, Berlin, Heidelberg, 1988, pp. 91–101.
[8] N Kalsbeek, J Richter, Preservation of burned bones: an investigation of the effects
of temperature and pH on hardness, Stud. Conserv. 51 (2) (2006) 123–138.
[9] R Neson, A microscopic comparison of fresh and burned bone, J. Foren. Sci. 37 (4)
(1992) 1055–1060.
[10] C Cattaneo, S DiMartino, S Scali, O Craig, M Grandi, R Sokol, Determining the
human origin of fragments of burnt bone: a comparative study of histological,
immunological and DNA techniques, Foren. Sci. Int. 102 (2-3) (1999) 181–191.
[11] S Grabherr, A Heinemann, H Vogel, G Rutty, B Morgan, K Woźniak, et al.,
Postmortem CT angiography compared with autopsy: a forensic multicenter study,
Radiology 288 (1) (2018) 270–276.
[12] C Lamm, M Dockner, B Pospischek, E Winter, B Patzak, M Pretterklieber, et al.,
Micro-CT analyses of historical bone samples presenting with osteomyelitis, Skelet.
Radiol. 44 (10) (2015) 1507–1514.
[13] A Thomsen, A Jurik, L Uhrenholt, A Vesterby, An alternative approach to
computerized tomography (CT) in forensic pathology, Foren. Sci. Int. 183 (1-3)
(2009) 87–90.
[14] P Bedford, Routine CT scan combined with preliminary examination as a new
method in determining the need for autopsy, Foren. Sci. Med. Pathol. 8 (4) (2012)
390–394.
[15] P Leth, The use of CT scanning in forensic autopsy, Foren. Sci. Med. Pathol. 3 (1)
(2007) 65–69.
[16] J Kuhn, S Goldstein, L Feldkamp, R Goulet, G Jesion, Evaluation of a
microcomputed tomography system to study trabecular bone structure, J. Orthop.
Res. 8 (6) (1990) 833–842.
[17] S Ellingham, M Sandholzer, Determining volumetric shrinkage trends of burnt
bone using micro-CT, J. Foren. Sci. 65 (1) (2020) 196–199.
[18] M Sandholzer, A Walmsley, P Lumley, G Landini, Radiologic evaluation of heat-
induced shrinkage and shape preservation of human teeth using micro-CT,
J. Foren. Radiol. Imaging 1 (3) (2013) 107–111.
Forensic Imaging 24 (2021) 200437
[19] I Velzen, M Raveendran, J Gonzalez-Rodriguez, Predictive models as screening
tools for DNA recovery from baked and burned porcine bones, Austin J. Foren. Sci.
Criminol. 2 (3) (2015) 1029.
[20] K Imaizumi, K Taniguchi, Y Ogawa, DNA survival and physical and histological
properties of heat-induced alterations in burnt bones, Int. J. Legal Med. 128 (2014)
439–446.
[21] T Tsuchimochi, M Iwasa, Y Maeno, H Koyama, H Inoue, I Isobe, et al., Chelating
resin-based extraction of DNA from dental pulp and sex determination from
incinerated teeth with Y-chromosomal alphoid repeat and short tandem repeats,
Am. J. Foren. Med. Pathol. 23 (3) (2002) 268–271.
[22] Y Zhang, Z He, S Fan, K He, C Li, Automatic thresholding of micro-CT trabecular
bone images, in: Proceedings of the International Conference on BioMedical
Engineering and Informatics, 2008, pp. 23–27.
[23] L Feldkamp, S Goldstein, M Parfitt, G Jesion, M Kleerekoper, The direct
examination of three-dimensional bone architecture in vitro by computed
tomography, J. Bone Miner. Res. 4 (1) (1989) 3–11.
[24] Morphometric parameters measured by Skyscan™ CT analyser software, Bruker
MicroCT, Available at https://library.westernsydney.edu.au/main/sites/default
/files/pdf/cite_Harvard.pdf, Accessed 21 January, 2021.
[25] A Parfitt, M Drezner, F Glorieux, J Kanis, H Malluche, P Meunier, et al., Bone
histomorphometry: standardization of nomenclature, symbols and units, J. Bone
Miner. Res. 2 (6) (1987) 595–610.
[26] C Trueman, K Privat, J Field, Why do crystallinity values fail to predict the extent
of diagenetic alteration of bone mineral? Palaeogeogr. Palaeoclimatol. Palaeoecol.
266 (3-4) (2008) 160–167.
[27] G Grupe, S Hummel, Trace element studies on experimentally cremated bone. I.
Alteration of the chemical composition at high temperatures, J. Archaeol. Sci. 18
(2) (1991) 177–186.
[28] C Smith, O Craig, R Prigodich, C Nielsen-Marsh, M Jans, C Vermeer, et al.,
Diagenesis and survival of osteocalcin in archaeological bone, J. Archaeol. Sci. 32
(1) (2005) 105–113.
[29] S Roberts, C Smith, A Millard, M Collins, The taphonomy of cooked bone:
characterizing boiling and its physico–chemical effects, Archaeometry 44 (3)
(2002) 485–494.
[30] C Nielsen-Marsh, R Hedges, Patterns of diagenesis in bone I: the effects of site
environments, J. Archaeol. Sci. 27 (12) (2000) 1139–1150.
[31] M Cummaudo, A Cappella, F Giacomini, C Raffone, N Màrquez-Grant, C Cattaneo,
Histomorphometric analysis of osteocyte lacunae in human and pig: exploring its
potential for species discrimination, Int. J. Legal Med. 133 (3) (2019) 711–718.
[32] T Vaughan, C McCarthy, L McNamara, A three-scale finite element investigation
into the effects of tissue mineralisation and lamellar organisation in human cortical
and trabecular bone, J. Mech. Behav. Biomed. Mater. 12 (2012) 50–62.
[33] S Qiu, D Fyhrie, S Palnitkar, D Rao, Histomorphometric assessment of haversian
canal and osteocyte lacunae in different-sized osteons in human rib, Anat. Rec. A
Discov. Mol. Cell. Evol. Biol. 272 (2) (2003) 520–525.
[34] C Chadefaux, I Reiche, Archaeological bone from macro-to nanoscale: heat-
induced modifications at low temperatures, J. Nano Res. 8 (2009) 157–172.
[35] M Marques, A Mamede, A Vassalo, C Makhoul, E Cunha, D Gonçalves, Heat-
induced bone diagenesis probed by vibrational spectroscopy, Sci. Rep. 8 (1) (2018)
1–13.
[36] S Danilchenko, A Koropov, I Protsenko, B Sulkio-Cleff, L Sukhodub, Thermal
behaviour of biogenic apatite crystals in bone: An X-ray diffraction study, Cryst.
Res. Technol. J. Exp. Ind. Crystallogr. 41 (3) (2006) 268–275.
[37] Mckinnon, M, Henneberg, M, Simpson, E & Higgins, D 2020, ‘A comparison of
crystal structure in fresh, burned and archaic bone–implications for forensic
sampling’, Foren. Sci. Int., pp. 110328.
[38] T Thompson, Recent advances in the study of burned bone and their implications
for forensic anthropology, Foren. Sci. Int. 146 (2004) S203–S205.
[39] D Cvetković, S Nikolić, V Brković, V Živković, Hyperostosis frontalis interna as an
age-related phenomenon–differences between males and females and possible use
in identification, Sci. Justice 59 (2) (2019) 172–176.
[40] M Brickley, Cribra orbitalia and porotic hyperostosis: a biological approach to
diagnosis, Am. J. Phys. Anthropol. 167 (4) (2018) 896–902.
[41] F Rühli, M Henneberg, Are hyperostosis frontalis interna and leptin linked? A
hypothetical approach about hormonal influence on human microevolution, Med.
Hypotheses 58 (5) (2002) 378–381.
[42] G Grévin, P Bailet, G Quatrehomme, A Ollier, Anatomical reconstruction of
fragments of burned human bones: a necessary means for forensic identification,
Foren. Sci. Int. 96 (2-3) (1998) 129–134.
[43] G Piga, A Malgosa, T Thompson, S Enzo, A new calibration of the XRD technique
for the study of archaeological burned human remains, J. Archaeol. Sci. 35 (8)
(2008) 2171–2178.
[44] A Bartsiokas, A Middleton, Characterization and dating of recent and fossil bone by
X-ray diffraction, J. Archaeol. Sci. 19 (1) (1992) 63–72.
[45] P Shipman, G Foster, M Schoeninger, Burnt bones and teeth: an experimental study
of color, morphology, crystal structure and shrinkage, J. Archaeol. Sci. 11 (4)
(1984) 307–325.
73
Chapter 6: General Discussion and

Conclusions

74
6.1 Discussion

Calcified tissues such as bone and teeth are inherently resilient, meaning they have great potential for

providing antemortem information and postmortem information, even long after death has occurred.

The skeleton has become a staple of forensic identification as it can provide a wealth of

anthropological evidence on factors such as age, sex and stature of unidentified remains. Additionally,

DNA profiles can often be obtained from skeletal material even after other tissues have been lost.

Although significant research has focused on determining identification from skeletal remains, the

impact of extreme trauma on these tissues remains relatively understudied. Incineration of remains is

a common occurrence in DVI situations, motor vehicle accidents, and household and industrial fires

and understanding how this might affect typical procedures such as DNA analysis and anthropological

profiling is vital to achieving a successful outcome. The comprehensive review presented in the

second chapter of this thesis highlights how little this is understood and suggests some areas of focus

for future research. While compiling this review it became apparent that no established protocols exist

for extracting DNA from incinerated bones, likely due to the fact that the chemical and compositional

differences between burned and unburned bone, beyond a decrease in organic content, is not fully

understood (Harbeck et al., 2011, Kalsbeek & Richter, 2006, Collins et al., 2002). Some studies have

suggested that incinerated bone is structurally similar to archaic bone, implying that ancient DNA

(aDNA) extraction techniques could be beneficial for forensic investigation of burned samples (Piga

et al., 2009, Stiner et al., 1995). As DNA is considered a primary method of scientific identification of

unknown human remains the main focus of my first study, chapter 3b, was to trial aDNA extraction

techniques on burned bone.

Before DNA extraction methods could be compared some issues with creating a “realistic”

incineration scenario were faced. I struggled to gain access to a suitable furnace that was both of

sufficient size for intact bones and capable of reaching and holding different temperatures at a

controlled ramp-up speed. As an alternative I opted for an open fire scenario, using only naturally

occurring bush material as fuel. At first, not knowing the temperature of the fire which the samples

were burned in seemed like a major limitation. However, on further consideration I decided that this

75
was a more realistic representation of a true incineration scenario. Ultimately, I decided to trade-off an

unrealistic but controlled experiment for a realistic environment where some variables might not be

controlled for. Another reason controlling temperature was not vital was that the primary focus of this

pilot study was to test different demineralisation methods on bones that were either fresh or

incinerated, not the effects of burning at different temperatures. I also wanted to test demineralisation

methods on bone that had undergone differing severities of burning as I expected this to have an

impact on bone integrity. For this reason, I included different durations of burning as an alternative to

controlled temperature. This made the study more robust, allowing me to investigate if burn severity

would impact the success of different demineralisation methods, a step advocated in aDNA methods.

Although, as expected, I discovered DNA yield decreased with increased time of exposure even after

120 minutes of burning in an open fire DNA could still be recovered. Although much of the pre-

existing research in this field focuses on the effects of burn temperature rather than durations, a

notable study by Grela et al., (2021) investigated multiple burn durations at a fixed temperature (400

°C) and achieved similar results. The primary finding of my study, however, was not the differences

between burn durations but the effectiveness of modified extraction techniques. Burned bone did not

respond especially well to ancient bone extraction protocols, in fact using less conservative methods

resulted in greater DNA yield so the ideal extraction method was still to be determined. Grela et al.,

2021 compared the effectiveness of four commercial extraction kits on both mitochondrial and

nuclear DNA recovery from bone and teeth. They showed that whilst bone provided more DNA than

teeth, heat negatively impacted the stability of mitochondrial DNA in both of these hard tissues. This

study (and others) also highlighted that both bone and teeth are demanding tissues to work with. Other

techniques to overcome the demands of incinerated tissues should be considered. These could include

free silica suspension, which is particularly useful for large sample volumes possessing low quantities

of DNA (Rohland et al., 2010), and direct PCR where samples are not isolated or purified to prevent

the loss of small quantities of DNA (Mercier et al., 1990).

76
The use of aDNA extraction methods on incinerated modern bone was suggested based on the

premise that their crystallite structures were similar. My extraction study, presented in chapter 3b, did

not support this, suggesting that the differences between ancient and burned modern bone are more

complex than just organic content. Hence, I decided to investigate further what actually changes

within the bone through the use of XRPD and SEM analysis. Whilst the focus of chapter 3 was to

investigate the impact of incineration on the outcome of DNA analysis, chapter 4 investigated the

possible mechanisms behind this outcome. The overall directive was twofold; could I use XRPD and

SEM to observe structural/compositional changes at different temperature intervals and if so, would

these heat-induced changes bear resemblance to ancient bones (as suggested by the literature) or are

they more like fresh modern bones (as suggested by results shown in chapter 3). Unlike the previous

study I was able to access an electric furnace with digital temperature controls, although a similar

issue with fitting intact samples was faced. The overall conclusion of this investigation was that not

all burned bones are equal; incineration at different temperatures results in a spectrum of changes with

little differences observed between samples burned below 600 °C, after which recrystallisation is

dramatic. This suggests that simply applying separate techniques for “burned” versus “unburned”

bones will likely be ineffectual for many samples. This could also indicate that whilst aDNA

extraction techniques investigated in the previous study may have been ineffectual, they may still

benefit some burned samples exposed to higher temperatures. Comparing the success of extraction

techniques in bones burned above and below this 600 °C threshold would be an interesting subject for

future research.

In regards to comparing archaic and incinerated samples, no ascertainable difference in crystalline

profile or appearance was observed. Despite these results, there were obvious differences between

these samples, for example the dense consistency of archaic bones made them significantly harder to

cut than fresh bones. It is more likely that although archaic bone and modern bone may have similar

crystalline profiles and microscopic appearance at low temperatures, there are many other differences

that cannot be observed using XRPD or SEM imaging. This could simply be differences in porosity,

collagen content or water content that could result in radically different responses to DNA extraction

77
methods. To expand on this further I considered other technologies that could be used to clarify the

relationship between incineration and tissue destruction, and histology seemed like the next logical

step. Moving forward, the results of chapter 4 were particularly useful in narrowing the focus of this

next study by eliminating some temperature intervals where little or no change was observed in bone

structure. As a PhD project is heavily time constrained, this was a very useful finding.

For my final study I first tried examining histological variation between unburned and burned bone

samples. In addition to examining heat driven compositional changes in bone that could affect DNA

recovery, I was also interested in linking these changes to observable effects of burning. However,

histology proved to be impossible on samples burned at higher temperatures, and no useful results

were obtained (Appendix i.). This ended up being a positive experience as it highlighted a need for

non-destructive methods of viewing changes in the whole bone. In some forensic contexts samples

may be so degraded that attempting to analyse them is a waste of resources (as I experienced first-

hand), however this is not always clear by simple visual appraisal. Having a method of evaluating

bone integrity without destruction of precious samples would be invaluable. Although much was

learned from XRPD and SEM, both these methods were limited by the need for some destruction of

the sample and inability to examine the whole bone. Methods such as synchrotron high resolution

imaging (or synchrotron-computed micro-CT SAXS/WAXD spectra (Ma et al., 2016)) can be used to

conduct 3D ex-vivo reconstruction of complex internal bone structure down to the molecular level.

Although this would be an ideal method of investigating structural changes of incinerated bones, the

current astronomical cost and logistical constraints make it unlikely that this avenue of research will

be pursued in the foreseeable future. More affordable and accessible technology such as Fourier

transform infrared and Raman spectroscopy have been used to provide quantitative assessment of

incinerated bone, however this provides very little in the way of visual assessment (Thompson et al.,

2011, Morris & Mandair, 2011). A more realistic and comprehensive approach for my research was

MicroCT scanning, which I used for my final study.

Chapter 5 demonstrated that differences in bone microstructure between burned and unburned

samples can clearly be observed using MicroCT without the need for any sample destruction.

78
Interestingly, although the overall decrease in bone porosity roughly correlates with the increased

crystallinity shown in chapter 4 there were additional fluctuations in porosity (primarily at 500 °C)

only observed using MicroCT. This is likely due to the much larger scale of MicroCT, which

emphasises the need for multiple analytical techniques to assess incinerated bone on different levels.

Ultimately, assessing burn temperature and duration is somewhat redundant in a practical context as

these variables are not known in a real event. By comparison, linking compositional changes to

observable effects of incineration could be vital for development of sample triaging. By consolidating

the changes in crystallite size observed in chapter 4 with visual changes observed using MicroCT and

SEM imaging, I was able to form a more comprehensive overview of the impact of incineration.

Although standard medical CT scanning is routinely used in many forensic laboratories, the higher

resolution MicroCT scanners needed to examine tissue integrity are largely limited by sample housing

capacity and cost. However, this knowledge is still valuable from a research perspective, and may be

particularly useful in the future as these technologies become more accessible and affordable.

Bone is an excellent source of discriminating information, however its behaviour under nonoptimal

conditions requires a greater understanding. The results of this thesis showed that not all burned bones

are equal and should be treated as such. Incineration of bone tissue is a complex process and therefore

requires multiple approaches for a comprehensive understanding. All of the techniques explored in

this thesis had advantages and disadvantages (Fig 6.1.), with some offering more information than

others at the expense of accessibility or cost. Knowing what each of these techniques can offer when

investigating incinerated remains is one of the primary findings of this study. There is a close

relationship between research and application, as each informs the other. For this reason, although not

all methods explored are applicable in a practical context, it is vital that they are understood for

further development. Even as the usage of MicroCT is limited now by factors such as sample housing

volume and cost, technological advancements could see these issues overcome in the near future.

When this happens, knowing exactly how this can be used for analysis is essential. Additionally,

knowing not only how these techniques can be used in cases of incineration, but also understanding

the impact of burning itself is vital. The other primary finding of this study demonstrated how burning

79
at different severities affects DNA recovery and bone integrity. Developing cohesive triaging tools

that could use this information to increase the likelihood of successfully recovering information from

burned bone would be an excellent future direction.

Fig 6.1. Summary table comparing the major advantages and disadvantages of
various analytical techniques. Each objective is identified as optimal,
nonoptimal or conditional – conditional meaning that the objective may be
somewhat achieved by use of a particular technique.

80
6.2 References

Collins, M, Nielsen–Marsh, C, Hiller, J, Smith, C, Roberts, J, Prigodich, R 2002, ‘The survival of

organic matter in bone: a review’, Archaeometry, vol. 44, no. 3, pp. 383-394.

Grela, M, Jakubczak, A, Kowalczyk, M, Listos, P & Gryzińska, M 2021, ‘Effectiveness of various

methods of DNA isolation from bones and teeth of animals exposed to high temperature’, Journal of

Forensic and Legal Medicine, vol. 78, pp. 102131.

Harbeck, M, Schleuder, R, Schneider, J, Wiechmann, I, Schmahl, W & Grupe, G 2011, ‘Research

potential and limitations of trace analyses of cremated remains’, Forensic Science International, vol.

204, no. 1-3, pp. 191-200.

Kalsbeek, N & Richter, J 2006, ‘Preservation of burned bones: an investigation of the effects of

temperature and pH on hardness’, Studies in Conservation, vol. 51, no. 2, pp. 123-138.

Ma, S, Boughton, O, Karunaratne, A, Jin, A, Cobb, J, Hansen, U & Abel, R 2016, ‘Synchrotron

imaging assessment of bone quality’, Clinical Reviews in Bone and Mineral Metabolism, vol. 14, no.

3, pp.150-160.

Mercier, B, Gaucher, C, Feugeas, O & Mazurier, C 1990, ‘Direct PCR from whole blood, without

DNA extraction’, Nucleic Acids Research, vol. 18, no. 19, pp. 5908.

Morris, M & Mandair, G 2011, ‘Raman assessment of bone quality’, Clinical Orthopaedics and

Related Research®, vol. 469, no. 8, pp. 2160-2169.

Piga, G, Thompson, T, Malgosa, A & Enzo, S 2009, ‘The potential of X‐ray diffraction in the analysis

of burned remains from forensic contexts’, Journal of Forensic Sciences, vol. 54, no. 3, pp. 534-539.

Rohland, N, Siedel, H & Hofreiter, M 2010, ‘A rapid column‐based ancient DNA extraction method

for increased sample throughput’, Molecular Ecology Resources, vol. 10, no. 4, pp. 677-683.

Stiner, M, Kuhn, S, Weiner, S & Bar-Yosef, O 1995, ‘Differential burning, recrystallization, and

fragmentation of archaeological bone’, Journal of Archaeological Science, vol. 22, no. 2, pp.223-237.

81
Thompson, T, Islam, M, Piduru, K & Marcel, A 2011, ‘An investigation into the internal and external

variables acting on crystallinity index using Fourier Transform Infrared Spectroscopy on unaltered

and burned bone’, Palaeogeography, Palaeoclimatology, Palaeoecology, vol. 299, no. 1-2, pp. 168-

174.

82
Appendix

83
Appendix i. Histological analysis of incinerated bone

The histology of incinerated bone compared to archaic samples was initially going to form another

chapter of this thesis. To achieve this, decalcification was performed on a range of burned and archaic

samples using the following method:

1. Small rectangular bone samples (~5x10 mm) were collected from larger sections.

2. 50 g of ethylenediaminetetracetic acid (EDTA) was dissolved in 4525 ml distilled water.

3. Sodium hydroxide (~25 g) added to EDTA solution until neutral pH 7.0 achieved.

4. Samples placed in individual test tubes and EDTA solution added.

5. Tubes placed on shaker plate set to low speed to gently agitate samples overnight

6. Each day-old solution was pipetted off, and fresh solution added.

7. Samples regularly x-rayed for signs of decalcification, frequency of x-rays was dependent on

the current speed of decalcification.

Following a month of EDTA incubation the incinerated samples would not decalcify, and subsequent

cutting, mounting and staining were unsuccessful. Even after prolonged decalcification, incinerated

samples were still so dense that attempts at cutting resulted in destruction of the sample and

significant damage to microtome blades. Comparatively partial success was achieved with some

archaic remains, specifically those taken from permafrost bison bone (Gold Run dig site, ~10-20 Ka),

and human burial sites at St Mary’s Cemetery (>100 years), Metaponto (~2.2-2.6 Ka) and Pompeii

(1.9 Ka). Even these archaic samples did not completely decalcify and after 28 days x-rays indicated

that decalcification would likely not progress further (Fig i.1.). The presence of calcified regions, as

well as the already degraded nature of the tissue resulted in severe tearing of most tissue samples (Fig.

i.2).

84
Fig i.1. X-rays of archaic bone samples decalcified in EDTA for a period of 28 days. At

approximately 19 days samples developed a ring of decalcified tissue around the calcified centre. By

28 days no further changes of the calcified centre were observed, so decalcification was halted.

Waiting for samples to decalcify only to find most were not viable was disappointing, particularly as

the focus on incinerated bone could not be further examined using this method. However, this did

provide some useful information about archaic samples and direction for future work. Although too

few samples are presented here to provide any statistical significance, it appears that the age of the

samples impacted the success of decalcification and tissue preparation (fig i.2). Although the much

more recent St Mary’s sample presented the most intact osteons, the permafrost bison sample from the

Gold Run site was very well preserved considering its advanced age. This supports the X-ray

diffraction analysis presented in Chapter 4, which concludes that the permafrost bison bone is not

significantly different in structure to modern bovine samples. It is also interesting to note that the

histological appearance of intact human osteons is not dramatically different from bovine osteons.

Following unsuccessful decalcification and histological analysis of incinerated bones, vacuum-based

resin infiltration was considered as a possible alternative to mounting. Resin embedding removes the

need for decalcification by infiltrating bone pores and reinforcing stability, which then allows

extremely dense and/or brittle incinerated material to be polished down, exposing the interior without

loss of structural integrity (Nielsen & Maiboe, 2000, Erben, R 1997). Preparing bone in this way

allows for in situ examination of the interior morphology in an unaltered state. This technique is often

used on dense archaic bone (Turner-Walker & Jans, 2008, Yang et al., 2003), and it was theorised that

this may be equally effective for incinerated material. Test samples burned at 100 °C and 700 °C were

prepared using EpoFix cold resin kit (© 2021 Electron Microscopy Sciences). Once embedded in

epoxy resin, samples were polished down to expose the interior. Embedded samples were then

85
examined using SEM with semi-satisfactory results (fig 6.4). Although this is a useful technique to

explore then interior structure of bone, higher quality images were obtained from the powdered

samples presented in Chapter 4.

The histomorphology of incinerated bone is something that could be further explored – I only trialled

one method of embedding, and this did not present significantly better results than un-embedded

samples. Although Methylmethacrylate (MMA) is the most commonly used medium for resin

embedding of undecalcified samples (Sterchi & Eurell, 1995, Buijs & Dogterom, 1983), other

commercially available resin kits may actually provide more reliable results without use of hazardous

chemicals (Mohsin et al., 2006). None of these resin embedding techniques have been properly

optimised for burned samples, and exploring this further could be of significant value for forensic

analysis.

86
Fig i.2. Histological sections of archaic bone stained with H&E (A-D) and Masson’s Trichrome (E-H). Although the H&E staining appeared to work slightly better than

Masson’s Trichrome on degraded tissue, the two stains were not dramatically different. The St. Mary’s samples (A & E) presented the most intact tissue and osteons, whilst

samples from Metaponto (C & G) and Pompeii (D & H) did not completely decalcify, resulting in severe tearing of the tissue and a lack of intact osteons. The permafrost Gold

Run samples (B & F) were the intermediary, showing some tearing and partially intact osteons.

87
Fig 6.4. SEM captures of resin embedded sample at low to high magnification. Figure A) shows a comparison sample burned at 100 °C, which is significantly more porous than the

incinerated samples presented in B) - D). The red arrow represents a region of interest on the same sample examined at higher magnification in figure C) & D). The granulated

appearance of bone at higher magnification C) is comparable to SEM captures presented in Chapter 4, however this was only observed within small pockets of tissue.

88
References

Buijs, R & Dogterom, A 1983, ‘An improved method for embedding hard tissue in poeymethyl

methacrylate’, Stain Technology, vol. 58, vol. 3, pp. 135-141.

Erben, R 1997, ‘Embedding of bone samples in methylmethacrylate: an improved method suitable for

bone histomorphometry, histochemistry, and immunohistochemistry’, Journal of Histochemistry &

Cytochemistry, vol. 45, no. 2, pp. 307-313.

Mohsin, S, O’Brien, F & Lee, T 2006, ‘New embedding medium for sectioning undecalcified bone’,

Biotechnic & Histochemistry, vol. 81, no. 2-3, pp. 99-103.

Nielsen, J & Maiboe, J 2000, ‘Epofix and vacuum: an easy method to make casts of hard substrates’,

Palaeontologia Electronica, vol. 3, no. 1, pp. 10.

Sterchi, D & Eurell, J 1995, ‘An evaluation of methylmethacrylate mixtures for hard tissue

embedding’, Journal of Histotechnology, vol. 18, 1, pp.45-49.

Turner-Walker, G & Jans, M 2008, ‘Reconstructing taphonomic histories using histological analysis’,

Palaeogeography, Palaeoclimatology, Palaeoecology, vol. 266, no. 3-4, pp. 227-235.

Yang, R, Davies, C, Archer, C, & Richards, R 2003, ‘Immunohistochemistry of matrix markers in

Technovit 9100 New-embedded undecalcified bone sections’, European Cells and Materials, vol. 6,

pp. 57-71.

89
Appendix ii. Awards and other achievements

Presentations

• 2021 “Recovering DNA from burned bone” ANZFSS SA branch award presentations, oral

presentation via zoom.

• 2020 “Comparing crystal structure in burned remains – implications for forensic

sampling” Florey postgraduate research conference. Poster presentation via zoom.

• 2018 “Can DNA be extracted from burned bones: analysis of various sampling sites in pig

and human models” Research presentation given at Forensic Science South Australia. Oral

presentation.

• 2018 “Growth patterns and individual variation in mid-sagittal facial soft tissue depth from

childhood to adulthood” 87th Annual Meeting of the American Association of Physical

Anthropologists. Austin, Texas, USA. Poster Presentation.

Teaching/lectures

• As a teaching assistant in Anthropological Anatomy II, I demonstrations during practical

exercises, help grade student assessments and provided the following lectures:

➢ 2018 “Skeletal Identification” Oral presentation.

➢ 2018 “Ecosensitivity of Human Biological Characters” Oral presentation.

➢ 2018 “DNA extraction from severely damaged bones: an analysis of various

sampling sites and extraction methods” Oral presentation.

Collaborations/professional acknowledgements

• Forensic Science South Australia (FSSA) – In addition to allowing me to present my

proposed research at the beginning of 2018, FSSA allowed me to complete a three month

contract working as mortuary technician whilst COVID-19 prevented me from working on

my thesis.

90
 • Advanced DNA, Identification & Forensic Facility (ADIFF) – Staff at the ADIFF provided

technical expertise, some consumables and access to low-copy DNA laboratory space.

• Faculty of Engineering, Computer and Mathematical Sciences, Adelaide University – Staff

provided training, technical expertise and use of x-ray diffractometer.

• Adelaide Microscopy – Staff at Adelaide Microscopy provided training and access for use of

MicroCT, SEM & EDX equipment, also assisted with interpretation of results and technical

expertise.

• Australian New Zealand Forensic Science Society (ANZFSS) – ANZFSS provides a lecture

series dedicated to developments in forensic science, which is directly related to my area of

research. Becoming a committee member has allowed me to network with experts in the field.

• SA Museum – Anna Russo of SA museum allowed me to assist Ellie Simpson in auditing the

museum’s stored skeletal collection, which was invaluable to the development of my

anthropological profiling skills.

Other achievements

• Publication of honors work: Mckinnon, M, Simpson, E & Henneberg, M 2018, ‘Growth

Patterns and Individual Variation in Mid‐sagittal Facial Soft Tissue Depth from Childhood to

Adulthood’, Journal of Forensic Sciences, vol. 63, no. 6, pp. 1641-1651.

• Ross Vining Memorial Student Scholarship – I was awarded this scholarship for the abstract I

submitted for the 22nd Triennial Meeting of the International Association of Forensic

Sciences. The conference was to be held in Sydney, 2021, and the scholarship was to assist

with the cost of registration, airfares and/or accommodation. However, the event was

cancelled due to COVID-19.

91